13 Red Light Therapy Benefits and Possible Risks

Updated on August 24, 2026 | Estimated reading time: 13 minutes

A device that looks like a glowing red panel now appears in homes, clinics, and sports-recovery settings around the world, and people use light-based devices for goals ranging from skin appearance to localized pain and wound care.

The 13 Red Light Therapy Benefits and Possible Risks discussed below do not arise from one fully settled mechanism, and they are not supported equally. Red and near-infrared photons may interact with cytochrome c oxidase and other photoacceptors, influencing mitochondrial signaling, nitric oxide, calcium, ATP, and reactive oxygen species in ways that depend on the cell's starting condition and the delivered dose. Human research supports some indication-specific outcomes, but a 2025 umbrella review found that certainty was moderate for only a limited group of outcomes and low or very low for most others.

This article explains what the science supports, where evidence remains uncertain, and which situations require professional guidance. It is educational information, not a diagnosis or a substitute for medical treatment. Any device claim should be checked against the exact wavelength, measured output, dose, intended use, instructions for use, and model-specific regulatory status.

What is red light therapy and how did it develop?

The Development History of Red Light Therapy

Red light therapy (RLT), often discussed within the broader field of photobiomodulation (PBM), applies non-ionizing visible red or near-infrared optical radiation at parameters intended to produce photochemical or photophysical effects without ablation. Common research wavelengths fall roughly within 600–700 nm for visible red and 780–1,100 nm for near-infrared, but the label "red light therapy"does not define one dose or one clinical treatment. A source can also produce heat if its irradiance, distance, exposure duration, or thermal design is inappropriate.

The clinical history did not begin with a wound-healing experiment on astronauts. Low-power laser research dates to the 1960s, including Endre Mester's animal work. Decades later, NASA-supported teams first developed LEDs for plant growth in space. Researchers subsequently tested high-intensity red and near-infrared LEDs in oxygen-deprived wounds in rats and in animal cell cultures. NASA's own account says the technology was not ultimately used in space for this purpose. The accurate link is that NASA-supported LED research helped advance and legitimize medical-light research—not that faster wound healing was documented in astronauts.

That distinction matters. Before exploring the 13 red light therapy benefits and possible risks in detail, it helps to anchor expectations with a quick orientation:

1. Define the spectrum precisely. Visible red and near-infrared light have different absorption and scattering behavior, but a wavelength alone does not identify the tissue reached or guarantee a biological effect.
2. Quantify the delivered dose. Relevant variables include average irradiance at the treatment plane, exposure time, illuminated area, pulse pattern, beam uniformity, and radiant exposure. Raw electrical wattage is not a treatment dose.
3. Match the intended use. Evidence for one source geometry, body site, and condition cannot automatically be transferred to a different device or application.
4. Separate research hypotheses from care claims. Laboratory investigations of cells, microbes, or inflammatory pathways do not show that a consumer device prevents or treats infection in people.

Understanding this development is the clearest route to evaluating what RLT can—and cannot—reliably do.

How red light therapy works: the mechanism behind the benefits

Proposed cellular pathways involved in red and near-infrared photobiomodulation

Photobiomodulation is not explained by a single proven pathway. Cytochrome c oxidase is an important proposed mitochondrial photoacceptor, while reviews also discuss opsins, flavoproteins, water-related effects, and light- or heat-sensitive ion channels. Reported downstream responses include changes in ATP, nitric oxide, calcium signaling, gene expression, and inflammatory mediators. Reactive oxygen species may briefly rise as signaling molecules in normal cells yet fall in some oxidatively stressed models, so "RLT reduces oxidative stress"is not a universal description.

Why doesn't simply using a higher-wattage device give better results?

Because biological response depends on dose rather than box wattage. PBM literature frequently describes a biphasic response: too little exposure may do nothing, while excessive exposure may reduce the intended response and can add heat or erythema. Radiant exposure is commonly expressed in J/cm² and, for a continuous source with reasonably uniform output, is calculated from irradiance in W/cm² multiplied by time in seconds. Even that number is incomplete unless wavelength, pulse characteristics, beam area, distance, and measurement method are reported.

Does wavelength really change which tissue gets treated?

Yes, wavelength affects absorption and scattering, but tissue penetration is a gradient rather than an exact map. Visible red light is generally attenuated more superficially than near-infrared light; however, skin color, water, blood, fat, source geometry, contact, and power density all affect how much light reaches a target. It is therefore too definite to say that 660 nm treats only skin or that 850 nm reliably reaches a joint or bone at a therapeutic dose. A dual-wavelength design may be useful for a defined purpose, but it is not a universal professional standard, and a single-wavelength device is not inherently ineffective.

Treatment parameters—wavelength, spectral bandwidth, average irradiance at the treatment plane, radiant exposure, pulse structure, session schedule, and target condition—separate reproducible research from unsupported marketing.

13 red light therapy benefits: what the evidence actually shows

Neutral documentation of skin and exercise-related red light therapy research

Evidence is not uniform across these 13 categories. "Benefit"here means a reported or plausible outcome under investigation, not a guarantee that every red-light device will produce it. A 2025 umbrella review synthesized 15 meta-analyses, 204 randomized trials, and more than 9,000 participants. It found moderate-certainty evidence for only several outcome-and-condition pairings, including hair density in androgenetic alopecia, disability in knee osteoarthritis, fatigue in fibromyalgia, pain in burning mouth syndrome, and some cognitive outcomes; most other endpoints were supported by low- or very-low-certainty evidence.

Skin and cosmetic benefits (Benefits 1–4)

Benefit 1 — Collagen stimulation and anti-aging. Small controlled and split-face studies report improvements in skin texture, elasticity, or wrinkles after specific red or red-plus-near-infrared protocols. Laboratory findings also support fibroblast and extracellular-matrix responses. However, study size, device parameters, comparators, and outcome measures vary, so no single irradiance or 8–12 week schedule should be presented as a universal anti-aging protocol. The defensible conclusion is that skin rejuvenation is promising and protocol-specific, not guaranteed.

Benefit 2 — Acne reduction. The original claim that approximately 630 nm red light directly suppresses Cutibacterium acnes is misleading. Blue-violet light, commonly around 407–420 nm, more directly excites bacterial porphyrins; red light is generally used for proposed anti-inflammatory effects or in combined blue-red regimens. A Cochrane review of 71 randomized studies found substantial variation and rated evidence for many light-therapy comparisons as very low quality. Some light-based devices have FDA clearance for narrowly defined acne indications, but clearance applies to the exact device and intended use, not to "red light therapy"as a whole.

Benefit 3 — Wound healing and scar reduction. Cell and animal research supports effects on fibroblasts, re-epithelialization, inflammation, and repair signaling. Human results are encouraging in selected wound-care settings, and hospitals may use PBM under clinical protocols. However, a 2026 systematic review of LED PBM for chronic lower-limb wounds found therapeutic potential but concluded that small samples, heterogeneous dosimetry, and methodological limitations prevented definitive efficacy conclusions. RLT should not replace infection control, vascular assessment, pressure relief, debridement, or other standard wound care.

Benefit 4 — Skin conditions under investigation. PBM has been explored for inflammatory or repair-related skin conditions, but evidence for psoriasis, eczema, and rosacea is not sufficient to treat generic red-light exposure as first-line therapy. These diagnoses have different mechanisms and standard treatments. Anyone with a photosensitive eruption, uncertain lesion, or changing pigmented lesion should seek dermatologic assessment before exposure.

Musculoskeletal and recovery benefits (Benefits 5–8)

Benefit 5 — Muscle recovery and reduced delayed-onset muscle soreness (DOMS). Some localized PBM trials and earlier reviews report less soreness or improved recovery when parameters and timing are tightly controlled. The evidence should not be described as consistently positive across delivery methods. A 2025 systematic review of five whole-body PBM studies involving 105 physically active participants found no evidence of better exercise performance or recovery biomarkers, although sleep outcomes were suggested in two studies. Results from localized laser or LED treatment cannot automatically validate full-body panels.

Benefit 6 — Joint pain and inflammation. This is one of the more credible musculoskeletal areas, but the outcome must be stated precisely. The 2025 umbrella review reported moderate-certainty evidence for improvement in disability associated with knee osteoarthritis; it did not establish that every arthritis condition or every pain outcome responds. PBM may be considered an adjunct for a defined condition and protocol, not a replacement for diagnosis, exercise therapy, medication, or other indicated care.

Benefit 7 — Bone healing support. Animal studies provide biological plausibility, while human fracture evidence remains limited. A systematic review and meta-analysis of randomized trials rated the evidence low to very low and found possible improvements in pain and function, but it did not establish routine acceleration of radiographic fracture healing. PBM should therefore be described as investigational support rather than an established fracture treatment.

Benefit 8 — Peripheral circulation improvement. PBM can influence nitric-oxide signaling and has produced localized vascular or perfusion changes in experimental settings. That mechanism does not establish a durable, whole-body circulation benefit or show that RLT treats peripheral arterial disease. Claims should be limited to transient, local, dose-dependent effects unless a clinical trial directly supports the target population and outcome.

Neurological, mood, and other systemic benefits (Benefits 9–13)

Benefit 9 — Sleep quality. Early studies suggest that certain red-light or whole-body PBM protocols may influence sleep measures. A small 2012 trial in 20 female athletes used 670 nm red light and reported changes in sleep and serum melatonin; it did not establish a general effect of near-infrared light on both melatonin and cortisol. A 2025 whole-body review found possible sleep improvement but included only five studies overall. The evidence does not support a universal evening wavelength, preset, or daily schedule.

Benefit 10 — Mood and seasonal affective support. RLT should not be conflated with established bright-light therapy for seasonal affective disorder, which is designed around ocular light exposure and circadian timing. Transcranial PBM for depression and anxiety remains experimental. Reviews describe encouraging preliminary findings but call for larger randomized trials; consumers should not use it to replace assessment or treatment for depression, bipolar disorder, suicidality, or another psychiatric condition.

Benefit 11 — Hair growth. Low-level laser and LED therapy has randomized-trial and meta-analytic support for increasing hair density in androgenetic alopecia, and the 2025 umbrella review rated this outcome at moderate certainty. FDA clearance exists for specific devices and populations. Proposed mechanisms include effects on follicular metabolism and the hair cycle, but "direct stimulation of follicle stem cells"is not a settled clinical explanation. Pattern hair loss should also be distinguished from scarring alopecia, alopecia areata, nutritional deficiency, and endocrine causes.

Benefit 12 — Nasal and upper-respiratory care. Narrow-band intranasal red light has been studied in allergic rhinitis, including a small controlled study reporting symptom improvement. That is not evidence that red light prevents influenza, treats a viral or bacterial upper-respiratory infection, or replaces antihistamines, intranasal corticosteroids, vaccines, testing, or antimicrobial treatment when indicated. This category should be described as a limited, condition-specific research area—not a "flu season"wellness benefit.

Benefit 13 — Cognitive function and neuroprotection. Transcranial PBM has been investigated in cognitive impairment and several neurological conditions. The 2025 umbrella review found moderate-certainty evidence for a pooled cognitive outcome, but trials differed in diagnosis, wavelength, dose, device geometry, and cognitive tests. Light attenuation through scalp and skull is a major technical constraint. The evidence supports further indication-specific research; it does not yet justify broad claims of neuroprotection or cognitive enhancement in healthy people.

Possible risks and where red light therapy does NOT apply

Model-specific eye and skin safety checks during red light therapy

IEC 62471 is a photobiological-safety standard for lamps and lamp systems. It specifies exposure limits, measurement methods, and a classification scheme for hazards from electrically powered incoherent optical sources, including LEDs, from 200 to 3,000 nm. A useful safety conclusion requires the exact source, spectrum, radiance or irradiance, apparent source size, distance, and exposure duration—not the color name alone.

RLT is generally reported as well tolerated under studied protocols, but "non-ionizing"does not mean risk-free. Eye and skin hazards depend on dose and spectral region. The photochemical retinal hazard commonly called blue-light hazard is weighted toward shorter visible wavelengths; high-intensity visible or infrared sources may also create thermal hazards. It is therefore inaccurate to say that every 660 or 850 nm session causes cumulative photochemical retinal damage or that goggles are universally mandatory. Users should not stare into a bright source and should follow the exact instructions for eye closure, distance, exposure, and model-specific eyewear. People with retinal disease, recent eye surgery, photosensitivity, or ocular medication should ask an eye-care professional before use.

Here are the key risk categories every user should understand:

• Overexposure and heat. Excessive irradiance or duration can produce discomfort, erythema, or thermal injury and may move exposure outside the intended biological window. Stop if the skin becomes painful or unusually hot.
• Photosensitizing medicines and products. Some antibiotics, retinoids, oncology drugs, herbs, and topical agents can increase reactions to visible light. A pharmacist or clinician should review the exact drug and spectrum; risk cannot be inferred from a generic list alone.
• Known or suspected tumors. Do not deliberately irradiate a tumor outside a clinician-directed protocol. PBM is not an anticancer treatment. This does not mean PBM has no role anywhere in oncology: evidence-based clinical guidelines support specific professional protocols for preventing or treating oral mucositis associated with some cancer therapies.
• Pregnancy. Fetal safety has not been adequately studied. Clinical services may use precautions such as avoiding abdominal exposure, but consumers should obtain obstetric guidance rather than assuming either universal safety or universal prohibition.
• Recent procedures, injections, or wounds. An open wound is not automatically a contraindication—wound care is a clinical PBM research area—but unsupervised exposure can delay appropriate diagnosis or compromise infection-control procedures. Follow the treating clinician's plan after surgery, injection, or wound treatment.
• Pulsed modes and photosensitive epilepsy. Visible flicker can trigger seizures in susceptible people. They should avoid pulsed or visibly flickering sources unless a qualified clinician and the device instructions confirm an appropriate protocol.
• Implanted electronic devices. Optical radiation does not inherently interfere with every implant, but heat, accessories, electrical safety, and the underlying condition may matter. Follow the implant and light-device instructions and seek clinician or manufacturer guidance when either identifies a restriction.

RLT is not established as a treatment for internal-organ disease, acute infection requiring standard medical care, or cancer itself. It should never delay antibiotics, antivirals, cancer therapy, wound assessment, or urgent evaluation. Its use as supportive care in selected oncology settings is a separate, clinician-directed application and should not be misrepresented as tumor treatment.

Are at-home devices a safe and reasonable option?

Evidence-based comparison of professionally supervised and at-home photobiomodulation

Is buying a home red light therapy device actually worth it?

It can be reasonable for a narrowly defined goal when a user chooses a well-documented device, follows its instructions, and accepts that evidence is indication- and device-specific. It is not possible to answer "yes"for all skin, recovery, pain, or medical goals. Home and professionally used devices overlap in output; the meaningful differences are intended use, clinical supervision, treatment geometry, safety controls, and whether the protocol matches the evidence.

The absence of supervision matters most when symptoms need diagnosis or when errors in target, dose, medication screening, or infection control could cause harm. Cosmetic use on intact skin is not equivalent to treating a chronic wound, neurological disorder, eye condition, or persistent joint pain.

What should a first-time buyer check before purchasing?

Four questions cut through most of the noise:

1. What is the exact intended use and regulatory status of this model? Verify an FDA clearance, authorization, or approval in the appropriate database when one is claimed; establishment registration alone is not enough. For an EU claim, request the Declaration of Conformity and any required notified-body details rather than relying on a CE logo.
2. Are optical measurements reproducible? Look for wavelength or spectrum, average irradiance at a stated distance and treatment plane, coverage uniformity, measurement method and instrument, and the resulting dose—not box wattage or a single peak reading at the lens.
3. Do the instructions define safe use? The document should specify distance, duration, frequency, eye precautions, heat limits, cleaning, contraindications, and what to do after an adverse reaction.
4. Can the supplier trace the exact model and support it? Confirm model-specific test reports, labeling, warranty, complaint handling, and after-sales support rather than accepting company-wide logos as proof for every item.

These checks apply to personal buyers, clinics, and commercial purchasers. For sourcing relationships, quality-system records, complaint handling, traceability, change control, and market-specific documentation matter alongside ordinary business terms. None substitutes for verifying the exact device's output and intended-use evidence.

A home device may provide a convenient way to reproduce a studied protocol, but convenience does not make it clinically equivalent to supervised care. Use the exact instructions rather than a generic "10–20 minutes daily"rule.

Is red light therapy scientifically proven? Weighing the evidence honestly

Evidence-certainty spectrum for red light therapy benefits and possible risks

The honest answer is that red light therapy occupies a wide evidence spectrum and cannot be labeled proven or unproven as one indivisible treatment.

The 2025 umbrella review is a useful high-level check: among 35 endpoints across 15 conditions, it identified significant results for 12 outcomes and moderate-certainty evidence for only five outcome categories. Most other outcomes were low or very low certainty because of heterogeneity, small-study effects, or both. This does not mean that every indication outside those five is ineffective; it means the available evidence cannot support uniformly confident claims.

Regulatory language must also stay precise. FDA establishment registration means an establishment is entered in a database and does not denote approval, clearance, authorization, or endorsement. A 510(k) clearance generally means FDA found a device substantially equivalent to a legally marketed predicate for a defined intended use after reviewing the required information. Supporting data can be nonclinical, clinical, or both, depending on the device. Clearance is meaningful, but it is not the same as premarket approval and does not prove every broader marketing claim. Anyone evaluating the 13 red light therapy benefits and possible risks should verify the exact device and indication rather than relying on an "FDA registered"label.

Common study limitations include small samples, inconsistent wavelength and dose reporting, short follow-up, multiple outcomes, and difficult blinding. A glowing active source can be hard to mimic convincingly, and sham designs vary. Differences in age, diagnosis, skin characteristics, treatment area, source geometry, and outcome definitions make direct comparisons unreliable.

For independent assessment, start with systematic reviews, clinical guidelines, PubMed-indexed trials, official regulatory databases, and recognized safety standards—not a product landing page. Check whether the study used the same body site, condition, source type, wavelength, dose, and schedule being marketed.

The fair verdict: PBM has credible, condition-specific clinical evidence and several legitimate medical applications, but many popular claims remain uncertain. Infection prevention, broad immune support, hormonal regulation, generalized detoxification, and cancer treatment should not be presented as established RLT benefits.

One practical note on the manufacturing side: REDDOT LED's reported ISO 13485:2016 system and model- or scope-specific CE, FCC, ETL, RoHS, and FDA records can matter for OEM/ODM quality and market-access review. They do not replace an exact model's intended-use authorization, photobiological-safety assessment, or indication-specific clinical evidence. Keeping that boundary clear protects manufacturers, brand clients, clinicians, and end users.

Key Takeaways

Red light therapy and photobiomodulation use non-ionizing visible red or near-infrared optical radiation, but wavelength alone does not define a treatment. Proposed mechanisms include cytochrome c oxidase and other photoacceptors, with context- and dose-dependent effects on ATP, nitric oxide, calcium, reactive oxygen species, and inflammatory signaling. The strongest recent synthesis found moderate-certainty evidence for a limited set of condition-specific outcomes and low or very low certainty for most others. Safe evaluation requires the exact intended use, spectrum, average irradiance at the treatment plane, radiant exposure, instructions, eye and thermal precautions, and model-specific regulatory documentation. More exposure is not necessarily better, and RLT should not delay established medical care.

FAQ

Is there any danger in using red light therapy?

Yes, although adverse effects in studied protocols are usually mild and transient. Excessive optical or thermal exposure can harm skin or eyes, and photosensitizing medicines, tumors, pregnancy, recent procedures, photosensitive epilepsy, and some eye conditions require added caution. The type of eye hazard depends on wavelength, radiance, source size, distance, and time; it is not accurate to label all 660 nm and 850 nm devices as causing the same photochemical injury. Do not stare into a bright source, and follow the exact model’s instructions and eye-protection requirements. IEC 62471 provides a framework for evaluating incoherent sources such as LEDs but does not make every device automatically safe.

Can red light therapy age your skin?

Studied low-level red-light protocols have generally investigated skin improvement rather than accelerated aging, but it is too broad to say that only UV can contribute to photoaging. UV is the dominant established cause, while visible and infrared radiation can also affect pigmentation, oxidative signaling, matrix metalloproteinases, and skin temperature under some exposure conditions. A non-thermal PBM protocol is different from chronic intense solar infrared or overheating. Follow the prescribed dose and stop if a device produces persistent redness, pain, or excessive heat.

What happens if I use red light therapy every day?

There is no universal daily schedule. Trials range from multiple sessions per week to other condition-specific regimens, and the same exposure time can deliver very different doses on different devices. For a continuous source, radiant exposure is approximately average irradiance multiplied by time, but distance, coverage, pulsing, and uniformity also matter. Daily use is appropriate only when the exact instructions or clinician-directed protocol call for it. Do not increase time simply because no immediate sensation occurs.

Is it scientifically proven that red light therapy works?

Some applications have randomized trials and systematic reviews, but the certainty varies by condition and outcome. A 2025 umbrella review found moderate-certainty evidence for selected outcomes, including androgenetic-alopecia hair density and knee-osteoarthritis disability, while most of the 35 assessed endpoints were supported by low or very low certainty. Wound care, skin rejuvenation, muscle recovery, sleep, mood, and neurological uses should each be judged from their own evidence and protocol. "Proven for everything"overstates the science; "unsupported in every use"also ignores genuine condition-specific evidence.

References

• Son Y, et al. Effects of photobiomodulation on multiple health outcomes: an umbrella review of randomized clinical trials. Systematic Reviews. 2025;14:160.
• Hamblin MR. Mechanisms and applications of the anti-inflammatory effects of photobiomodulation. AIMS Biophysics. 2017;4(3):337–361.
• NASA Spinoff. NASA Research Illuminates Medical Uses of Light. 2022.
• Cochrane. The use of light as a therapy for acne. Cochrane Review CD007917.
• Wunsch A, Matuschka K. A controlled trial to determine the efficacy of red and near-infrared light treatment in patient satisfaction, reduction of fine lines, wrinkles, skin roughness, and intradermal collagen density increase. Photomedicine and Laser Surgery. 2014;32(2):93–100.
• Ng JNC, Wanitphakdeedecha R, Yan C. Efficacy of home-use light-emitting diode device at 637 and 854 nm for facial rejuvenation: a split-face pilot study. Journal of Cosmetic Dermatology. 2020;19(9):2288–2294.
• Miranda MB, et al. Clinical dosimetry and efficacy of LED photobiomodulation for chronic lower-limb wound healing: a systematic review of randomized trials. Lasers in Medical Science. 2026;41(1):117.
• Álvarez-Martínez M, Borden G. A systematic review on whole-body photobiomodulation for exercise performance and recovery. Lasers in Medical Science. 2025;40:55.
• Neto FCJ, et al. Effects of photobiomodulation in the treatment of fractures: a systematic review and meta-analysis of randomized clinical trials. Lasers in Medical Science. 2020;35(3):513–522.
• Vital de Azevedo WM, et al. Photobiomodulation and sleep quality: systematic review and meta-analysis. Lasers in Medical Science. 2026;41(1):122.
• Zhao J, et al. Red light and the sleep quality and endurance performance of Chinese female basketball players. Journal of Athletic Training. 2012;47(6):673–678.
• Neuman I, Finkelstein Y. Narrow-band red light phototherapy in perennial allergic rhinitis and nasal polyposis. Annals of Allergy, Asthma & Immunology. 1997;78(4):399–406.
• Lee TL, Ding Z, Chan AS. Can transcranial photobiomodulation improve cognitive function? A systematic review of human studies. Ageing Research Reviews. 2023;83:101786.
• Kashiwagi S, et al. Photobiomodulation and nitric oxide signaling. Nitric Oxide. 2023;130:58–68.
• Darwin E, et al. Review of transcranial photobiomodulation for major depressive disorder. Brain Stimulation. 2016;9(5):641–647.
• Lueangarun S, et al. A systematic review and meta-analysis of randomized controlled trials of home-use low-level light/laser therapy devices for pattern hair loss. Journal of Clinical and Aesthetic Dermatology. 2021;14(11):E64–E75.
• Zadik Y, et al. Systematic review of photobiomodulation for the management of oral mucositis in cancer patients and clinical practice guidelines. Supportive Care in Cancer. 2019;27:3969–3983.
• Memorial Sloan Kettering Cancer Center. About Your Photobiomodulation Therapy.
• International Electrotechnical Commission. IEC 62471:2006—Photobiological safety of lamps and lamp systems.
• International Commission on Non-Ionizing Radiation Protection. Visible Radiation and Infrared Radiation.
• U.S. Food and Drug Administration. Important Reminders about Registration and Listing.
• U.S. Food and Drug Administration. Premarket Notification 510(k).
• European Commission. CE marking.
• International Organization for Standardization. ISO 13485:2016—Medical devices—Quality management systems—Requirements for regulatory purposes.
• Cho S, et al. Effects of infrared radiation and heat on human skin aging in vivo. Journal of Investigative Dermatology Symposium Proceedings. 2009;14(1):15–19.