What Does a TENS Machine Actually Do?

A TENS machine is a battery-powered box the size of a deck of cards that sends a mild electrical current into your skin through sticky pads. It costs less than a single physiotherapy appointment, needs no prescription, and has almost no side effects. It is also one of the most argued-over objects in pain care, because the people who love theirs and the guideline committees who have reviewed it are, at this point, saying opposite things.

The honest summary is narrower than the packaging. TENS reliably takes the edge off pain while it is switched on. The evidence that anything is different after you peel the pads off is weak. And the clinical guidelines that have looked at it for specific conditions have mostly come back saying don’t bother — which sounds damning until you notice they were answering a different question from the one you are asking when you plug it in.

What the machine is actually doing

TENS stands for transcutaneous electrical nerve stimulation: through the skin, electrical, nerve. Two or four self-adhesive electrodes go on or around the sore area, wired to the unit. You turn the intensity up until you feel a strong buzzing or tingling that stops short of being unpleasant, and leave it there — twenty minutes, an hour, sometimes longer.

What you are stimulating is not the painful structure. The current is far too weak to reach a worn joint or a compressed nerve root. It preferentially excites the large, fast, well-insulated nerve fibres just under the skin, the ones that carry touch and vibration, because those are the easiest fibres in the body to switch on. They arrive at the same segment of spinal cord as the small, slow fibres carrying pain signals, and heavy traffic on the large fibres suppresses what the small ones get through.

That is the gate control theory Melzack and Wall published in 1965, and it is the same reason rubbing a banged elbow helps. A TENS machine is a rubbed elbow you do not have to keep rubbing.

There is a second mechanism, and it is why the machine has a frequency dial rather than just a volume knob. Stimulation also triggers the spinal cord and brainstem to release the body’s own opioid peptides — and, in animal work, high and low frequencies do it through different receptors. High frequency acts at delta opioid receptors; low frequency at mu receptors, the ones morphine acts on. That distinction is not trivia. It is the reason the two settings feel different, last for different lengths of time, and suit different people.

The two settings that matter

Every machine has more programmes than anyone needs — “massage”, “knead”, “tap”, numbered presets nobody documents. Underneath, almost all of them are variations on two things.

Two practical points follow from that table. The first is that intensity matters more than the programme number: across the research, the sessions that worked were the ones turned up to a strong but comfortable sensation, not a polite tickle. If you cannot clearly feel it, you are unlikely to be getting anything from it.

The second is speculative but worth raising with whoever prescribes for you. In animals, low-frequency TENS stops working in morphine-tolerant animals, because it is leaning on the same receptors. If you take opioids daily, that is a reason to give the high-frequency setting a fair trial before concluding the machine does nothing for you.

Where the evidence stands

While it is switched on

This is the part that is settled. The largest synthesis in the field pooled 381 randomised trials covering 24,532 people. Restricted to the trials that compared real TENS against a dummy machine and measured pain during or immediately after stimulation — 91 trials, 4,841 participants — pain intensity was lower with the real thing by a standardised effect of −0.96, which is a large effect, and the reviewers rated the certainty as moderate rather than low.

Across all 381 trials there were no serious adverse events. For a treatment with a mains-free battery and a fifty-dollar price tag, that is a genuinely good safety record.

After you take the pads off

Here it falls apart. An overview of every Cochrane review of TENS in chronic pain found the reviews themselves were well conducted and the evidence inside them was very low quality — small trials, high risk of bias, inconsistent methods. The authors could not confidently say whether TENS helps or harms chronic pain, disability, quality of life, medication use, or how people rate their own change.

That is not a finding that TENS fails. It is a finding that thirty years of small, badly designed studies have left the question open, which is a different and more frustrating result.

Fibromyalgia — the best single trial in the field

The exception is a properly powered trial in 301 women with fibromyalgia, randomised to active TENS, placebo TENS or none, using it two hours a day during activity for four weeks at modulated frequency and the highest tolerable intensity.

Movement-evoked pain fell by one point more on a ten-point scale than with placebo, and fatigue by 1.4 points. A single point sounds thin. The global figures are more striking: 70% of the active group said they had improved, against 31% on placebo and 9% on nothing at all. Fewer than 5% had any adverse effect, and none serious.

Two things about that design are worth keeping. Participants used it during activity, not lying down afterwards. And the outcome was pain on moving, not pain at rest — which is the pain that actually stops people doing things.

Neuropathic pain

A Cochrane review of TENS for neuropathic pain pooled five sham-controlled studies totalling 207 people and found pain about 1.6 points lower on a ten-point scale — but graded the evidence very low quality and declined to draw a conclusion. If you have numb or burning feet, there is also a safety wrinkle covered further down, and it is not a small one.

Low back pain — where the guidelines said no

Two heavyweight bodies have now reviewed TENS for back pain and recommended against it. The World Health Organization’s 2023 guideline on chronic low back pain says it should not be part of routine care. Britain’s NICE guidance on chronic pain says not to offer it for chronic primary pain, and its separate back pain guideline says the same.

Knee osteoarthritis

Pooled trials favour TENS, sometimes strikingly. They also show the lopsided pattern that appears when small positive studies get published and small negative ones quietly do not. Treat the enthusiastic numbers in this corner of the literature with more suspicion than the modest ones elsewhere.

Why the guidelines and the meta-analysis disagree

They are not measuring the same thing, and once you see it the contradiction mostly dissolves.

The meta-analysis asked: does pain go down while the machine is running, compared with a machine that produces no sensation? Answer: yes, clearly. The guidelines asked: does adding TENS to a person’s care leave them measurably better off weeks or months later, at a cost the system should carry? Answer: nobody has shown that, so we cannot recommend it.

Both answers are correct. What a guideline cannot easily say is “this may be worth fifty dollars of your own money for an hour of relief in the evening,” because that is not the kind of judgement guidelines are built to make.

There is also a problem nobody has solved: you cannot blind someone to being buzzed. A placebo TENS unit that produces no sensation announces itself immediately. Every trial in this field is contaminated by that, which is exactly why the certainty ratings stay stubbornly low no matter how many studies get added.

The part the manual doesn’t mention

It stops working if you use it the same way every day. In a study of 100 adults given twenty minutes of TENS daily for five days, the high-frequency effect had measurably faded by day four and the low-frequency effect by day five. This is tolerance at the same opioid receptors the machine recruits — the mechanism that makes it work is the mechanism that wears out.

Animal work suggests that alternating or mixing frequencies delays it, and the fibromyalgia trial above used a modulated frequency rather than a fixed one, which may be part of why it held up over four weeks. The practical version: use it for something specific rather than leaving it on all day, vary the setting, and if a machine that used to help has stopped, try a week off before deciding it was never working.

What it does not do

It does not treat the cause. Nothing about a surface current slows joint erosion, reverses cartilage loss, widens a narrowed spinal canal or regrows a damaged nerve. It manages a symptom, and it manages it for as long as it is on.

It is not a muscle stimulator, though most units sold today combine both and the distinction is buried in the manual. EMS drives a motor nerve to make a muscle contract, for rehabilitation after an injury or surgery. TENS aims at sensory nerves. Buying a combined unit is fine; expecting the TENS side to build strength is not.

And it does not improve circulation, flush anything, or reduce inflammation in any way that has been demonstrated. Those claims live on the packaging and nowhere else.

Using one safely

Some situations need a conversation with your own clinician rather than a rule of thumb:

• A pacemaker, implanted defibrillator or any other active implant. The device is the reason to ask, not the distance between it and the pads. This is a decision for the cardiologist who manages it.
• Numb skin. If you have peripheral neuropathy, the sensation you are told to turn up to is the same sensation that would warn you the current is too high. Burns under electrodes on skin that cannot report them are a documented problem, not a theoretical one.
• Pregnancy. Not over the abdomen or lower back. Purpose-built TENS units are used in labour, but that is a different decision made with a maternity team.
• Epilepsy, or an undiagnosed pain. A new pain wants a diagnosis before it gets a machine — masking it is the one way an object this benign can cause harm.
• Driving, or operating anything. A pad peeling loose delivers a sharp jolt. Nobody needs that at a highway merge.

Beyond that, the common problems are dull: itchy red squares where the gel sat, pads that stop sticking, a nine-volt battery dying mid-session. Moving the pads a centimetre between sessions and washing the skin first solves most of it.

Buying one in Canada

TENS units are Class II medical devices here, which means the manufacturer must hold a Health Canada licence to sell one — but you do not need a prescription to buy it. You can check any specific model against the Medical Devices Active Licence Listing before ordering, and it is worth thirty seconds, because plenty of what turns up on marketplace listings is not licensed for sale in Canada at all.

Prices run from about forty dollars for a basic two-channel unit to a few hundred for a clinical-grade one with more programmes than you will use. The meaningful difference between them is not the number of presets; it is whether the intensity adjusts in fine enough steps to find a strong-but-comfortable level, and whether the controls are usable by hands that hurt. Buttons you can find without your glasses beat an extra twelve modes.

The running cost is the electrode pads. They lose their grip after a few dozen applications and a replacement set is a few dollars every month or two — modest, but it never stops, so factor it in.

On coverage: no provincial health plan pays for one. Many private extended health plans will, usually with a prescription or a letter from a physiotherapist, so ask before buying rather than after. The federal Non-Insured Health Benefits program for First Nations and Inuit clients explicitly excludes electrotherapy devices, so there is no appeal route there.

Before spending anything, ask a physiotherapist to try one on you. Most clinics have units, most will let you feel what a properly set-up session is like, and finding the pad placement that works for your particular pain is the single thing most likely to decide whether the machine ends up in a drawer.

Giving it a fair test

Two weeks is enough to know. Use it during something that hurts to do — the walk, the dishes, the drive — rather than as a reward afterwards. Start on the high-frequency setting, turned up to a strong tingle, and give it twenty to thirty minutes. Move the pads if it does nothing; placement is guesswork at first and small shifts change everything. Rate the pain before, during, and an hour after.

Then judge it on the during-and-just-after numbers, because that is the honest claim TENS makes. If you were hoping the pain would be gone the next morning, no trial in the literature supports that expectation, and holding it is the fastest way to conclude the thing is useless when it was doing exactly what it does.

It is a volume knob, not a cure. The mistake is not buying one — it is expecting it to still be turned down after you have taken it off.

So, what does it actually do?

It makes a painful hour less painful, for as long as it is on and a while after, at very low risk and very low cost. That is the whole of it, and it is not nothing. For fibromyalgia there is real trial evidence that using it during activity lets people move with less pain and less fatigue, which is worth more than the modest numbers suggest. For chronic low back pain the guideline bodies have looked and declined to recommend it, and they are not wrong to.

Where it earns its place is alongside everything else that makes a painful body’s day more bearable — heat, pacing, movement you can sustain — rather than in place of whatever is actually treating the cause. If you are still working out what that looks like, living with chronic pain in Canada covers the wider ground, and Pain BC runs free programs for people anywhere in the country who want help building the rest of the plan.