Irradiance & Dosing: The Physics of Treatment

You can own the most expensive red light therapy panel on the market and still get zero benefit from it. Not because the device is faulty, but because you’re using the wrong dose. Dosing is the single most important variable in photobiomodulation — more important than wavelength, more important than device brand, and far more important than treatment time alone.

Yet dosing is also the most widely misunderstood aspect of red light therapy. Manufacturers quote irradiance numbers that sound impressive but often mean nothing in practice. Users treat for arbitrary durations based on forum advice. And almost nobody accounts for distance, the variable that changes everything.

This page explains the physics of dosing from first principles so you can calculate an accurate, effective dose every time.

The two numbers that matter

Red light therapy dosing rests on two measurable quantities:

Irradiance (mW/cm²)

Irradiance is the power density — how much optical power is delivered per unit area of tissue. It is measured in milliwatts per square centimetre (mW/cm²) or watts per square centimetre (W/cm²).

Think of irradiance as the “brightness” of the light falling on your skin at a specific point. A device might produce 100 mW/cm² at 0cm (touching the skin) but only 25 mW/cm² at 15cm distance.

Irradiance tells you how fast energy is being delivered. A higher irradiance delivers the target dose in less time.

Fluence (J/cm²)

Fluence is the total energy dose — how much energy has been delivered per unit area over the entire treatment session. It is measured in joules per square centimetre (J/cm²).

Fluence is what your cells actually “experience.” It is the cumulative result of irradiance applied over time. Two treatments with identical fluence will produce identical biological effects, regardless of whether the irradiance was high (short treatment) or low (long treatment) — within reason.

The relationship between them

The fundamental dosing equation is:

Fluence (J/cm²) = Irradiance (W/cm²) × Time (seconds)

Note the unit conversion: 1 W = 1000 mW, and 1 J = 1 W × 1 second.

So if your device delivers 50 mW/cm² (0.05 W/cm²) and you treat for 120 seconds:

Fluence = 0.05 × 120 = 6 J/cm²

This equation is deceptively simple, but it contains everything you need. If you know any two of the three variables (irradiance, time, fluence), you can calculate the third.

Rearranged for practical use

To find treatment time: Time (seconds) = Fluence (J/cm²) ÷ Irradiance (W/cm²)

Example: You want a 4 J/cm² dose and your device delivers 40 mW/cm² at your treatment distance. Time = 4 ÷ 0.04 = 100 seconds (1 minute 40 seconds)

To find the dose you’re getting: Fluence = Irradiance × Time

Example: You treat at 30 mW/cm² for 10 minutes (600 seconds). Fluence = 0.03 × 600 = 18 J/cm² — likely too high for most applications.

The inverse square law: why distance changes everything

Here is where most people get dosing wrong. Irradiance is not a fixed property of a device — it changes dramatically with distance.

The inverse square law states that the intensity of light from a point source decreases with the square of the distance from the source:

I₂ = I₁ × (d₁/d₂)²

Where:

I₁ = irradiance at distance d₁
I₂ = irradiance at distance d₂

For a device that measures 200 mW/cm² at the surface (0cm):

At 15cm: approximately 50–80 mW/cm² (reduced, but not perfectly following the inverse square law because LED panels are extended sources, not point sources)
At 30cm: approximately 20–40 mW/cm²
At 60cm: approximately 5–15 mW/cm²

The inverse square law applies strictly only to point sources. LED panels are extended sources — arrays of multiple LEDs spread across a surface. For extended sources, irradiance falls off more slowly at close distances (where the panel subtends a large solid angle) and approaches inverse-square behaviour at greater distances (where the panel approximates a point source).

The practical implication: doubling your distance from a panel does not quarter the irradiance (as the pure inverse square law predicts), but it does reduce it substantially. At typical treatment distances of 15–30cm, most panels deliver 25–60% of their surface irradiance.

What this means for your dose

If you calculate your treatment time based on the manufacturer’s stated irradiance (usually measured at 0cm), you will systematically underdose at any practical treatment distance.

Example:

Manufacturer states 200 mW/cm² (measured at surface)
You treat at 20cm distance, where actual irradiance is 60 mW/cm²
You calculate treatment time assuming 200 mW/cm²: for 6 J/cm², that’s 30 seconds
Actual dose delivered in 30 seconds at 60 mW/cm²: 0.06 × 30 = 1.8 J/cm² — less than a third of your intended dose

This is why the irradiance number on a spec sheet is almost meaningless without knowing the measurement distance.

Why manufacturer claims are often misleading

The red light therapy industry has a systemic honesty problem with irradiance claims. Here are the common ways numbers get inflated:

Measurement at 0cm (surface contact)

Most manufacturers measure irradiance with the sensor pressed directly against the LEDs. Nobody actually uses a panel this way — you stand 15–45cm away. Surface irradiance is physically the highest possible reading and tells you nothing about the dose you’ll actually receive.

Single-point measurement

Irradiance varies across the surface of a panel. The centre typically produces the highest reading (where beams from multiple LEDs overlap), whilst the edges produce lower readings. Quoting only the peak centre reading overstates the average irradiance across the treatment area.

Cherry-picking wavelengths

Some sensors are more responsive to certain wavelengths. A manufacturer might use a sensor that reads higher at 660nm and lower at 850nm (or vice versa), then quote whichever number is more impressive.

Ignoring angular distribution

LEDs emit light in a cone. The irradiance directly in front of the LED (at 0°) is much higher than at 30° or 60° off-axis. A panel’s effective treatment area depends on the beam angle of its LEDs — wide-angle LEDs distribute light more evenly but at lower peak intensity, whilst narrow-angle LEDs concentrate light into a smaller area with a higher peak reading.

How to get accurate numbers

The only reliable way to know your dose:

Buy a solar power meter — A basic handheld meter capable of reading mW/cm² costs £20–40. Position it at your actual treatment distance and measure.
Measure at your treatment distance — Not at the surface. At the distance you actually stand or sit from the panel during treatment.
Take multiple readings — Measure at the centre of the panel, at the edges, and at intermediate points. Average the readings for a more accurate irradiance value.
Remeasure periodically — LED output degrades over time (typically 5–10% per year). Annual recalibration ensures your dose calculations remain accurate.

Therapeutic dose ranges by condition

The published literature provides reasonably consistent dose ranges for different therapeutic applications. These are fluence values at the tissue surface — the dose actually reaching target cells will be lower due to tissue attenuation.

Skin conditions (anti-ageing, collagen, wound healing)

Optimal range: 3–6 J/cm²
Evidence: Wunsch and Matuschka (2014) used 30 J/cm² delivered over 30 minutes (relatively low irradiance) with significant improvements in collagen density. However, most wound healing studies use 1–6 J/cm² (Whelan et al., 2001; Minatel et al., 2009).
Frequency: 3–5 sessions per week
Risk of overdose: Low at these levels, but chronic overdosing may impair rather than enhance healing (see biphasic response below).

Joint pain and arthritis

Optimal range: 4–8 J/cm² at the skin surface
Evidence: Bjordal et al. (2003) found optimal doses of 4–8 J/cm² at the joint surface for knee osteoarthritis. Note that tissue attenuation means only a fraction of this reaches the joint capsule — which is why NIR wavelengths (which penetrate deeper) are preferred.
Frequency: 3–5 sessions per week for acute conditions; 2–3 per week for maintenance

Muscle recovery and performance

Optimal range: 3–6 J/cm² per muscle group
Evidence: Leal Junior et al. (2015) reviewed multiple trials and found that pre-exercise PBM at 1–6 J/cm² reduced muscle fatigue and post-exercise creatine kinase levels.
Timing: Before exercise for performance; after exercise for recovery. Pre-exercise application has stronger evidence.

Hair growth

Optimal range: 3–6 J/cm²
Evidence: Lanzafame et al. (2013) used 655nm at 4 J/cm² with significant increases in hair count. Kim et al. (2013) used similar parameters.
Frequency: Every other day to daily

Neurological applications (transcranial PBM)

Optimal range: 10–30 J/cm² at the scalp surface
Evidence: Higher surface doses are needed because the skull attenuates approximately 95% of incident light. Naeser et al. (2014) used 810nm at approximately 25 J/cm² per site.
Important: These are specialist applications. The high doses required reflect skull attenuation, not a need for more energy at the cellular level.

Oral mucositis

Optimal range: 1–4 J/cm²
Evidence: Bensadoun et al. (1999) used 2 J/cm² at 830nm. Bjordal et al. (2011) confirmed efficacy at 1–3 J/cm² across multiple trials.

The biphasic dose response

The biphasic dose response — also called the Arndt-Schulz curve — is the most important concept in PBM dosimetry. It states that:

Low doses produce a stimulatory (beneficial) effect
Higher doses produce diminishing returns
Excessive doses produce an inhibitory (harmful) effect

This follows a characteristic inverted U-shaped curve: benefit increases with dose up to an optimum, then decreases beyond it. At sufficiently high doses, the treatment becomes actively counterproductive.

Huang et al. (2009) provided the most comprehensive analysis of this phenomenon. Their findings:

Below 0.001 J/cm²: No measurable effect — insufficient photon density to trigger cellular responses
0.5–4 J/cm²: Optimal stimulation for most cell types and conditions
5–10 J/cm²: Still beneficial but with diminishing returns
Above 10–16 J/cm²: Inhibitory effects begin to appear in many cell types
Above 50 J/cm²: Actively damaging — can induce apoptosis (cell death) rather than stimulation

The mathematical basis

The biphasic response can be modelled as two competing processes:

Stimulatory process — Follows a saturating function (similar to Michaelis-Menten kinetics). At low doses, each additional photon adds meaningfully to the cellular response. At higher doses, the system approaches saturation — CCO is fully activated, and additional photons produce diminishing returns.
Inhibitory process — Excessive ROS generation beyond the cell’s antioxidant capacity leads to oxidative stress. This process has a higher threshold but, once activated, overwhelms the stimulatory effect.

The net biological effect is the sum of these two processes:

Net effect = Stimulation(dose) − Inhibition(dose)

At low doses, stimulation dominates. At high doses, inhibition dominates. The optimum sits where the gap between stimulation and inhibition is greatest.

Practical implications

The biphasic response means:

More is not better. Doubling your treatment time does not double the benefit — it may eliminate it entirely.
Undertreating is safer than overtreating. A dose of 2 J/cm² may not be optimal for a given condition, but it’s unlikely to be harmful. A dose of 30 J/cm² might actively impair healing.
Treatment frequency matters more than session dose. Five sessions of 4 J/cm² across a week is more effective than one session of 20 J/cm². Each session pushes the cell into its stimulatory range without crossing into inhibition.
Whole-body panels require shorter treatment times. Large panels deliver irradiance to a much larger tissue area. The total energy delivered is irradiance × area × time. More area means more total energy — and more potential for systemic effects.

Irradiance rate effects

Whilst fluence (total dose) is the primary determinant of biological effect, there is emerging evidence that irradiance rate (how fast the dose is delivered) also matters.

High-irradiance, short-duration treatments and low-irradiance, long-duration treatments may produce the same fluence but different biological outcomes. This is because:

Very high irradiance may generate ROS faster than the cell can process them, tipping the balance toward inhibitory effects even at a “correct” total dose.
Very low irradiance may not generate sufficient instantaneous ROS signalling to trigger the desired cellular cascade, even if the cumulative dose is adequate.

The practical range where this effect is minimal spans approximately 10–200 mW/cm² — covering the vast majority of consumer and clinical devices. Below 10 mW/cm², treatments become impractically long. Above 200 mW/cm², there is a theoretical risk of rate-dependent inhibition, though this is not well characterised in humans.

Hadis et al. (2016) reviewed irradiance effects and concluded that, within the 10–200 mW/cm² range, fluence is the dominant factor and irradiance rate effects are secondary.

Building a dosing protocol

Here is a step-by-step process for calculating an accurate treatment protocol:

Step 1: Identify your target condition

This determines your target fluence. Use the ranges above as a starting point:

Skin: 3–6 J/cm²
Joints: 4–8 J/cm²
Muscle: 3–6 J/cm²
Hair: 3–6 J/cm²

Step 2: Measure your device’s irradiance at treatment distance

Use a solar power meter at the distance you’ll actually treat. If you don’t have a meter, assume your irradiance is 25–40% of the manufacturer’s stated surface irradiance at a 15–20cm treatment distance.

Step 3: Calculate treatment time

Time (seconds) = Target fluence (J/cm²) ÷ Measured irradiance (W/cm²)

Example: Target 4 J/cm², measured irradiance 50 mW/cm² (0.05 W/cm²) Time = 4 ÷ 0.05 = 80 seconds per treatment zone

Step 4: Account for treatment area

If you’re treating multiple body zones (e.g., face, then chest, then back), calculate the time for each zone separately. The dose is per unit area — treating for longer because you’re covering more area doesn’t increase the dose at any single point. You simply spend more total time to cover all zones.

Step 5: Set frequency

For most conditions, 3–5 sessions per week is supported by the literature. Daily treatment is acceptable for many applications, provided the per-session dose stays within the optimal range. Rest days may allow cellular processes to complete before the next stimulatory input.

Common dosing mistakes

Using manufacturer’s surface irradiance for time calculations — Always measure at your actual treatment distance.
Treating for the same duration regardless of device — A panel producing 100 mW/cm² at treatment distance requires half the time of one producing 50 mW/cm². Time is not dose.
Assuming longer is better — The biphasic response means excessive treatment can reverse your gains.
Ignoring distance consistency — Moving closer or further during treatment changes your irradiance and therefore your dose. Maintain a consistent distance throughout.
Conflating total energy with fluence — Total energy (joules) depends on the treatment area. A small handheld device and a large panel might deliver the same fluence (J/cm²) but very different total energy. Fluence at the tissue surface is what determines the cellular response.
Not accounting for clothing or barriers — Any material between the device and your skin (clothing, glass, plastic) absorbs or reflects some light, reducing the effective irradiance. Treat on bare skin whenever possible.

A note on pulsed versus continuous wave

Some devices offer pulsed (PW) modes alongside continuous wave (CW). In pulsed mode, the LEDs switch on and off at a set frequency (commonly 10 Hz, 40 Hz, or 73 Hz).

For dose calculations with pulsed light, the effective irradiance is:

Effective irradiance = Peak irradiance × Duty cycle

If a device pulses at 50% duty cycle (on half the time), the effective irradiance is half the peak value. Treatment times must be doubled to achieve the same fluence as CW mode.

Some research suggests that specific pulse frequencies may have additional biological effects — particularly 10 Hz (alpha brainwave frequency) for neurological applications and 40 Hz (gamma frequency) for Alzheimer’s research (Iaccarino et al., 2016). However, for most therapeutic applications, CW mode is simpler and equally effective.

References

Bensadoun RJ et al. (1999). Low-energy He/Ne laser in the prevention of radiation-induced mucositis. Support Care Cancer, 7(4):244-252. PMID: 10423050
Bjordal JM et al. (2003). A systematic review of low level laser therapy with location-specific doses for pain from chronic joint disorders. Aust J Physiother, 49(2):107-116. PMID: 12775206
Bjordal JM et al. (2011). A systematic review with meta-analysis of the effect of low-level laser therapy in cancer therapy-induced oral mucositis. Support Care Cancer, 19(8):1069-1077. PMID: 21660670
Hadis MA et al. (2016). The dark art of light measurement: accurate radiometry in low-level light therapy. Lasers Med Sci, 31(4):789-809. PMID: 26964800
Huang YY et al. (2009). Biphasic dose response in low level light therapy. Dose Response, 7(4):358-383. PMID: 20011653
Iaccarino HF et al. (2016). Gamma frequency entrainment attenuates amyloid load and modifies microglia. Nature, 540(7632):230-235. PMID: 27929004
Kim H et al. (2013). Low-level light therapy for androgenetic alopecia: a 24-week, randomized, double-blind, sham device-controlled multicenter trial. Dermatol Surg, 39(8):1177-1183.
Lanzafame RJ et al. (2013). The growth of human scalp hair mediated by visible red light laser and LED sources. Lasers Surg Med, 45(8):487-495. PMID: 24078483
Leal Junior EC et al. (2015). Effect of phototherapy on exercise performance and markers of exercise recovery. Lasers Med Sci, 30(2):925-939. PMID: 24249354
Minatel DG et al. (2009). Phototherapy promotes healing of chronic diabetic leg ulcers. Photomed Laser Surg, 27(1):93-99. PMID: 19196110
Naeser MA et al. (2014). Significant improvements in cognitive performance post-transcranial, red/near-infrared LED treatments. Photomed Laser Surg, 32(2):115-126. PMID: 24568233
Whelan HT et al. (2001). Effect of NASA light-emitting diode irradiation on wound healing. J Clin Laser Med Surg, 19(6):305-314. PMID: 11776448
Wunsch A, Matuschka K. (2014). A controlled trial to determine the efficacy of red and near-infrared light treatment. Photomed Laser Surg, 32(2):93-100. PMID: 24286286