Opportunities for the utilisation of health technology
Opportunities for the utilisation of health technology
Introduction
In this report, I examine the commonly available Omron TENS device, which carries a CE0197 marking. First, I discuss relevant abbreviations and directives and explain what the CE0197 marking signifies. I then familiarize myself with the device’s operating principle and its associated patents. I also test the device on myself to consider whether using TENS could reduce the need for medication. Finally, I explore whether the device could be further developed or integrated with other technology, such as a computer or smartphone.
In order to understand how the device interacts with the human body, I highlight key aspects of the nervous system and musculature. I assume the reader has basic knowledge of human physiology, biology, and the nervous system, as well as a sufficient understanding of electrical engineering concepts such as alternating vs. direct current, voltage, resistance, and impedance.
On the scope of the report
It is natural to focus on health-sector technology that is easily accessible, safe to test, and requires no significant funding. I have two Omron E2 Elite TENS electronic nerve stimulators, both of which bear the CE0197 marking. The device is primarily used for pain relief, though it has other applications which I will not explore in detail here.
Abbreviations, directives and laws
To keep the report clear, I use generally accepted abbreviations and definitions. However, some terms—such as the CE marking—require more detailed explanation.
CE (European Conformity)
The CE marking indicates that a product complies with EU requirements and must be present on certain categories of products to be sold in the European Economic Area (EEA). The CE marking is a manufacturer’s declaration that the product meets all specified requirements for safety, health, environmental protection, and consumer protection. Compliance is overseen by various authorities across different ministerial branches, depending on the product.
Medical (MDD) CE marked devices
In Finland, the National Supervisory Authority for Welfare and Health (Valvira) oversees the CE compliance of medical devices. MDD refers to the Council of Europe’s Medical Devices Directive (93/42/EEC), which defines a “medical device” as follows (Linnavuori 2015):
“Medical device” means any instrument, apparatus, implement, software, material or other article, whether used alone or in combination, together with any accessories, including software designed by its manufacturer for diagnostic and/or therapeutic purposes and necessary for its proper functioning, which the manufacturer intends to be used for human beings for: – diagnosis, prevention, monitoring, treatment or alleviation of a disease; – diagnosis, monitoring, treatment, alleviation or compensation of an injury or disability; – examination, replacement or modification of the anatomy or of a physiological function; – regulation of conception; and which does not achieve its principal intended action in or on the human body by pharmacological, immunological or metabolic means, but which may be assisted in its function by such means.”
(Adapted from Council Directive 93/42/EEC)
CE Classification for Medical Devices and Omron E2 Elite CE0197
In Finland, Directive 93/42/EEC has been implemented via the Act on Medical Devices and Supplies 629/2010, which empowers Valvira to issue binding regulations. Equipment manufacturers demonstrate CE compliance by affixing the CE marking to their products. Depending on the risk category, the conformity assessment can be carried out by the manufacturer or, if required, by an external party (a Notified Body, NB). If an external evaluator is involved, this is indicated by a numeric code following the “CE” mark (for example, in CE0197, the code 0197 identifies the NB as TÜV Rheinland LGA Products GmbH, which evaluated the device).
There are four classes of medical devices: Class I (with subclasses Is for sterile devices and Im for measuring devices) – the lowest risk, Class IIa, Class IIb, and Class III – the highest risk. Devices are classified according to the level of risk their use poses to the user. Classification is determined using the rules in Annex IX of the MDD (93/42/EEC). In addition, the manufacturer (or authorized representative) must sign a Declaration of Conformity attesting that the product meets all requirements (Linnavuori 2015).
EUDAMED (2022) is a European database on medical devices where one can search for a Notified Body or evaluator by device manufacturer, device type code, or NB number (the sequence after “CE”). For example, CE0197 indicates NB 0197 is TÜV Rheinland LGA Products GmbH, the evaluator to whom the manufacturer declared conformity.
CERTIPEDIA (2022) contains certifications issued by TÜV Rheinland LGA Products GmbH. For example, the Omron E2 Elite model is not listed by name, but Omron Healthcare Co. Ltd appears under several system certifications (reflecting certified quality systems for multiple device classes).
CE certificates for a device can be obtained from the manufacturer and, in this case, also from the importer (Berner Oy). (Appendix 1 of the report provides the CE Declaration of Conformity for a newer model of the Omron TENS device, not included in this report.) All Omron TENS devices are classified as medical devices. The Omron E2 Elite in question is a Class IIa device (MDD Article 9, Annex IX, Regulation 9).
According to Regulation 9, Annex IX of the MDD (Council Directive 93/42/EEC), active therapeutic devices intended to produce or exchange energy are generally classified as Class IIa, unless they have characteristics that supply energy in a potentially dangerous way (by nature, intensity, or site of action), in which case they fall into Class IIb. Active devices for monitoring or directly influencing Class IIb therapeutic devices are themselves classified as Class IIb. This classification scheme is explained further in MDR-b (2022).
Article 9 of Annex IX also emphasizes that the manufacturer must have an audited quality system (per MDD Article 10) in place. This is why CERTIPEDIA (2022) lists Omron Healthcare only under system (quality) certifications. For a Class IIa device like CE0197, the Notified Body (TÜV Rheinland LGA Products GmbH) is entitled to review the quality system documentation, post-market studies, and other information at least annually. In addition, for Class IIa, the NB must be allowed to inspect technical documentation and samples, and it must perform an unannounced factory audit at least once every five years (MDR 2022).
Depolarization and polarization
Depolarization refers to a reduction in a cell’s membrane potential due to redistribution of charge. In a neuron, depolarization makes the inside of the cell less negative (moving the resting potential closer to zero). When a neuron depolarizes, its resting voltage (approximately –70 mV at rest) becomes less negative. The spread of depolarization along and between neurons constitutes the nerve impulse (action potential). During depolarization, positively charged sodium (Na⁺) ions flow into the cell, reducing the charge difference across the cell membrane.
Endorphins
Endorphins, often called endogenous morphines, are neurotransmitter hormones produced by the body that bind to opioid receptors, inducing pleasure and reducing pain perception. For example, β-endorphins are 18–33 times more potent than morphine in their analgesic effect (Loh, Tseng, Wei & Li 1976).
NNT, NNH and the placebo effect
A fundamental aspect of clinical studies is comparing treatment outcomes to a placebo. A placebo is an inactive treatment (e.g., a pill with no active drug, a sham surgery, or TENS with no electrical output) that can produce a psychological benefit known as the placebo effect.
Number Needed to Treat (NNT) is a metric indicating how many patients must be treated for one patient to benefit. For example, the NNT for the antidepressant vortioxetine (Brintellix) in clinical trials is about 7. This means that in those trials roughly 46–49% of patients on vortioxetine responded, compared to 34% on placebo; in other words, treating seven patients yields one additional responder compared to placebo. Similarly, about eleven patients must be treated for one additional remission.
Number Needed to Harm (NNH) indicates how many patients, on average, would experience an adverse effect from a treatment. In vortioxetine’s case, approximately 1 in 36 treated patients discontinued treatment due to side effects, so NNH ≈ 36 (Fimea 2014).
(In the above example, vortioxetine’s NNT=7 and NNH=36 were derived from clinical trial data: one additional responder per 7 treated, and one discontinuation due to side effects per 36 treated.)
Peripheral nervous system
The peripheral nervous system (PNS) comprises all nerves outside the brain and spinal cord, including the cranial nerves and spinal nerves. It has two major divisions: the autonomic nervous system (controlling involuntary functions of internal organs) and the somatic nervous system (controlling voluntary movements and sensory information). The autonomic nervous system is further divided into the sympathetic (active “fight-or-flight” responses) and parasympathetic (restful “rest-and-digest” responses) systems. Acetylcholine is the primary neurotransmitter of the parasympathetic system, while the sympathetic system primarily uses norepinephrine.
In the brain, a cluster of neuron cell bodies is called a nucleus, and a bundle of axons is called a pathway (or tract). In the PNS, a bundle of axons is called a nerve, and a cluster of cell bodies is called a ganglion. The central and peripheral nervous systems connect at the spinal cord and brainstem. The brainstem (comprising the midbrain, pons, and medulla oblongata) regulates vital functions like alertness and the sleep-wake cycle (e.g., via the suprachiasmatic nucleus). Eleven of the twelve cranial nerves emanate from the brainstem (the exception being the optic nerve, which connects directly to the brain). The cranial nerves mediate sensory and motor functions of the head and certain internal organs. Spinal nerves are organized into segments associated with dermatomes (skin regions). Bell-Magendie’s law states that the dorsal roots of spinal nerves carry incoming sensory information, while the ventral roots carry outgoing motor commands (Paavilainen 2016, 44–47; Hämäläinen 2017).
Frequencies in TENS devices
In TENS therapy, frequency refers to the number of electrical pulses per second, measured in hertz (Hz). For example, 30 pulses per second is 30 Hz. Frequency selection is significant because different frequencies can produce different therapeutic effects (e.g., high vs. low frequency stimulation in TENS).
TENS (transcutaneous electrical nerve stimulation)
Transcutaneous electrical nerve stimulation (TENS) is a form of electrotherapy in which low-voltage electrical current is applied to the skin via electrodes to stimulate nerves for therapeutic purposes. This technique can produce pain relief (analgesia) and is also known broadly as electrotherapy.
About the functioning of the human nervous system and musculature
Since the operation of a TENS device is related to human electrochemical functions, it makes sense to first examine the basics of how the human nervous system and musculature work. For clarity, I will simplify some aspects of nervous system function by describing them in terms of analog (graded) and binary (all-or-nothing) signals, even though in reality the physiology is more complex.
On the development and function of nerve cells and synapses
Neurons can be categorized as sensory neurons (which gather information from external or internal environments and are part of the PNS), motor neurons (which carry impulses from the central nervous system to muscles and glands), or interneurons (which connect neurons to each other and are the most numerous type). Interneurons form complex networks in which information processing occurs.
A neuron’s cell body (soma) contains the nucleus and genetic material that regulate its function. Extending from the soma are dendrites, which receive signals from other neurons. Each neuron usually has one axon, a long projection that transmits signals away from the cell body. An axon may branch into many terminals at its end. A single neuron can form connections (synapses) with tens of thousands of other neurons. Convergence refers to multiple neurons sending input to one neuron, whereas divergence refers to one neuron sending output to multiple other neurons.
Neurons form circuits; in the simplest reflex arc, for example, a sensory neuron synapses onto a motor neuron, which then activates an effector (such as a muscle). The cell bodies of sensory neurons reside outside the CNS in ganglia, while in the CNS clusters of cell bodies are nuclei. These nuclei and ganglia allow simple reflexes to be processed without involving the brain, preserving brain capacity for more complex tasks (Campbell 2011, 1131–1136).
Throughout development, neural connections (synapses) form and are pruned. Sensory and motor neurons develop early, guided by genetic instructions that establish connections within and between brain regions. Although one might assume synapse count only increases over a lifetime, in reality the number of synapses peaks and then decreases as redundant connections are pruned and neural circuits are optimized. Neurons that do not establish useful connections undergo programmed cell death (apoptosis). If a target cell does not provide sufficient trophic factors, the neuron will activate apoptosis and be removed. Conversely, when a connection is successfully made, the target cell releases growth factors that reinforce the synapse. It is estimated that about 70% of neural connections change daily; hence the adage “use it or lose it” applies – regular use of neural pathways helps maintain them, whereas inactivity leads to their weakening. As useful connections strengthen, neurons mature by developing insulating myelin sheaths around axons (analogous to wires being insulated), which increases the speed and efficiency of signal transmission. Notably, the amount of grey matter (unmyelinated neural tissue) in the brain tends to decrease with age as this pruning and optimization occur. For example, covering one eye will cause the brain to reduce connections serving that eye over time. This principle is even applied therapeutically (e.g., patching a stronger eye in a child with strabismus to strengthen neural connections for the weaker eye) (Penttonen 2017).
In other words, neurons extend axons and dendrites that seek out connections. Axonal growth is guided by a growth cone at the tip, influenced by contact cues and chemical signals (chemotaxis). If an axon’s filopodia successfully connect with a dendrite’s projections, a synapse forms—a process called synaptogenesis (Penttonen 2017; Paavilainen 2016, 55–68). Unsuccessful or excess connections are pruned away. Through these mechanisms, the developing nervous system refines itself, retaining efficient pathways and eliminating others.
Peripheral nervous system, sympathetic and parasympathetic
The peripheral nervous system can be functionally divided into the somatic nervous system (controlling voluntary movements and transmitting sensory information) and the autonomic nervous system (controlling involuntary organ functions). The somatic system includes efferent nerves to skeletal muscles and afferent nerves from sensory receptors. The autonomic system is split into the sympathetic division (engaged during stress or activity) and the parasympathetic division (engaged during rest and recovery). The sympathetic system (“fight or flight”) primarily uses norepinephrine as a neurotransmitter, while the parasympathetic (“rest and digest”) uses acetylcholine as its main neurotransmitter.
Neurons using the same neurotransmitter often form systems that regulate specific functions. For instance, the cholinergic system (acetylcholine) is involved in alertness and learning, the dopaminergic system (dopamine) in movement control and reward, the adrenergic system (adrenaline/norepinephrine) in arousal and emotion, and the serotonergic system (serotonin) in mood and sleep regulation.
Neurotransmitter effects are graded and transient—lasting only until the neurotransmitter is removed from the synaptic cleft (either by reuptake into nerve terminals or breakdown by enzymes). Each neuron integrates many inputs; some are excitatory, some inhibitory. The algebraic sum of these EPSPs (excitatory postsynaptic potentials) and IPSPs (inhibitory postsynaptic potentials) determines how the neuron’s membrane potential changes and whether it reaches the threshold to fire an action potential. If neurons were exclusively excitatory, neural activity could escalate uncontrollably. In practice, inhibitory signals (IPSPs) are crucial for moderating activity—for example, when focusing attention, IPSPs help suppress irrelevant inputs while EPSPs enhance relevant signals. Many neurons also have a baseline spontaneous firing rate, which is modulated up or down by incoming EPSPs and IPSPs (Paavilainen 2016, 41–42; 134–137).
Interneuronal networks can even generate spontaneous rhythmic activity. Most synapses occur between a presynaptic axon terminal and a postsynaptic dendrite. However, at the neuromuscular junction (a synapse between a motor neuron’s axon and a muscle fiber), the motor neuron releases acetylcholine to cause muscle contraction (Paavilainen 2016, 42).
Flow of information in a nerve cell
Information in the nervous system is transmitted using both binary and analog signals. The binary aspect, akin to digital communication, is seen in the all-or-nothing firing of neurons: a neuron is either at rest (not firing, representing “0”) or it fires an action potential (representing “1”). In a resting state, a neuron’s membrane potential is about –70 mV (inside negative relative to outside). When it fires, the membrane potential briefly goes to roughly +50 mV. Depolarization is a decrease in the voltage difference across the cell membrane (making the interior less negative), and hyperpolarization is an increase in that voltage difference (making the interior more negative). A depolarizing change (moving toward threshold) in a postsynaptic neuron is called an EPSP, while a hyperpolarizing change is an IPSP.
If a local depolarization reaches the trigger threshold (around –50 mV), an action potential is generated. The action potential is an all-or-nothing event: once threshold is reached, the nerve impulse produced is the same magnitude regardless of stimulus strength. (Hyperpolarizations, by contrast, do not trigger action potentials but make them less likely by moving the membrane potential further from threshold.) Stronger stimuli are encoded not by larger impulses (since impulse size is constant) but by higher frequency of action potentials.
During an action potential, voltage-gated sodium channels open, causing a rapid influx of Na⁺ and a reversal of the membrane polarity (the inside becomes positive relative to the outside). This depolarization at one segment of the axon triggers neighboring segments to depolarize, propagating the nerve impulse along the axon at speeds of about 1–10 m/s (for an unmyelinated fiber, depending on axon diameter). If the axon is wrapped in myelin sheath with gaps called nodes of Ranvier, the impulse leaps from node to node (saltatory conduction), reaching speeds up to ~120 m/s.
After firing, a neuron cannot immediately fire again until the membrane potential is restored. Sodium-potassium pumps (Na⁺/K⁺ ATPase) actively restore the resting ionic balance, a process that takes ~2 ms (the refractory period) and ensures one-way propagation of the impulse. Thus, a neuron can fire at most around 500 impulses per second. Because all action potentials are similar in magnitude, the nervous system uses frequency coding (how frequently impulses occur) to encode the intensity of a stimulus (Paavilainen 2016, 38–40).
Depolarization and hyperpolarization can be understood in terms of the neuron’s membrane potential, which exists because of differing ion concentrations across the cell’s semipermeable membrane. Typically, the cytoplasm is high in potassium (K⁺) and low in sodium (Na⁺), whereas the extracellular fluid is high in Na⁺ and low in K⁺. Inside the cell, negatively charged molecules (proteins, amino acids, sulfate, phosphate) serve as immobile anions. Ions cross the membrane either through selective ion channels (passively) or by active pumps. Large anionic proteins cannot cross and thus contribute to the negative interior. K⁺ tends to diffuse out of the cell (down its concentration gradient), making the inside more negative until an equilibrium is reached where the electrical pull of K⁺ back in balances its outward concentration gradient. This equilibrium defines the K⁺ equilibrium potential. However, Na⁺ also leaks inward (both its concentration and electrical gradients drive it into the cell), making the actual resting membrane potential slightly less negative than the K⁺ equilibrium potential. The Na⁺/K⁺ pump maintains these gradients by pumping Na⁺ out and K⁺ in, using ATP for energy (Campbell 2011, 1091–1097).
For example, glucose (blood sugar) oxidized in mitochondria releases energy to form ATP, which the Na⁺/K⁺ pump uses. ATP also mediates many other processes, such as phosphorylation of proteins during signaling. (Heino & Vuento 2008.)
While all cells have a membrane potential, only nerve and muscle cells can rapidly change their membrane potential (excitability). The resting potential is the normal membrane potential when a cell is not excited. Neurons have various gated ion channels that open in response to stimuli. In sensory neurons, the stimulus might be external (pressure, temperature, etc.). Depending on which ion channels open, a stimulus can cause a graded hyperpolarization or depolarization in the neuron. Such changes are called graded potentials because their amplitude depends on stimulus intensity. If a depolarization exceeds the firing threshold, an action potential ensues — a strong, self-propagating depolarization where voltage-gated Na⁺ channels drive the membrane potential positive. Hyperpolarizing graded potentials, on the other hand, inhibit action potentials by making the membrane more negative than the resting level.
Once initiated, an action potential’s amplitude is not proportional to the stimulus strength (it’s all-or-nothing); instead, a stronger stimulus causes multiple action potentials in succession (higher frequency). Action potentials are generated by voltage-sensitive ion channels that respond to changes in membrane voltage. During an action potential’s depolarization phase, the membrane potential reverses polarity (inside becomes positive relative to outside as Na⁺ floods in). Immediately after, Na⁺ channels inactivate and K⁺ channels help repolarize the membrane. After an impulse, the membrane potential is restored by Na⁺/K⁺ pumps and ion channel resets. Neurotransmitters released at synapses are then either taken back up into presynaptic terminals or broken down enzymatically to terminate the signal (Paavilainen 2016, 43).
Acceleration and slowing down of the nervous system
Neurotransmitters can be either excitatory or inhibitory. Excitatory neurotransmitters depolarize the postsynaptic membrane (bringing it closer to the ~–50 mV firing threshold), whereas inhibitory neurotransmitters hyperpolarize the membrane (further from threshold). A depolarizing change is an EPSP and a hyperpolarizing change an IPSP. The most common inhibitory neurotransmitter in the CNS is GABA (γ-aminobutyric acid), and the most common excitatory neurotransmitter is glutamate. Other important neurotransmitters include acetylcholine, serotonin, dopamine, and norepinephrine (Hämäläinen 2017).
Neurons in the brain that utilize the same neurotransmitter often form interconnected systems that regulate certain functions. For example, there are cholinergic pathways for arousal and learning (acetylcholine), dopaminergic pathways for motor control and reward (dopamine), adrenergic pathways for stress and alertness (norepinephrine), and serotonergic pathways for mood and sleep (serotonin). Neurotransmitter level changes are continuous (not stepwise) and their effects last only until the transmitter is removed from the synapse.
Because each neuron may receive many EPSPs and IPSPs simultaneously, the net effect on its membrane potential is the sum of these inputs. If the combined EPSPs minus IPSPs drive the neuron above threshold, it fires an action potential. In a balanced system, inhibitory neurons prevent runaway excitation. For instance, in focusing attention on a task, inhibitory signals suppress distractions while excitatory signals enhance the processing of the relevant stimulus. Many neurons also fire spontaneously at a baseline rate, which is adjusted up or down by incoming EPSPs and IPSPs to accelerate or decelerate neural activity (Paavilainen 2016, 41–42; 134–137).
Additionally, interneuron networks themselves can generate spontaneous impulses. Most synapses are axon-to-dendrite, but some are axon-to-muscle (neuromuscular junctions). At a neuromuscular junction, a motor neuron releases a neurotransmitter (acetylcholine) that causes muscle fibers to contract (Paavilainen 2016, 42).
About the function of the musculature
Movement is produced by muscle contraction, while relaxation of muscle (release of tension) occurs passively. Skeletal muscles that move a joint work in antagonistic pairs: for example, a flexor muscle bends a limb, and an extensor muscle straightens it. To move a limb in opposite directions, one muscle of the pair contracts while the other relaxes.
The ability of muscle to contract arises from its structure. Each skeletal muscle is composed of bundles of muscle fibers (muscle cells). Within each muscle fiber are many threadlike myofibrils. Myofibrils consist of repeating units called sarcomeres, which are the functional units of contraction. Each sarcomere contains myofilaments of two types: thick filaments made of myosin and thin filaments made of actin. The overlapping arrangement of actin and myosin filaments gives skeletal muscle its striated (striped) appearance under a microscope. The sarcomere is defined by Z-lines at its boundaries (Z-discs where actin filaments attach). The central region of the sarcomere with only myosin filaments is the H-zone. The region with only actin filaments is the I-band, and the length of myosin filaments (including overlap with actin) defines the A-band (Campbell 2011, 1153–1161).
In skeletal muscle tissue, connective tissue layers organize the fibers: each muscle fiber is wrapped in endomysium; bundles of fibers (fascicles) are wrapped in perimysium; and the entire muscle is wrapped in epimysium. Muscle cells are multinucleated and filled with myofibrils. The cytoplasm of a muscle cell is called sarcoplasm, and the cell membrane is the sarcolemma. The myofibrils’ actin and myosin filaments, aligned in register, create the striation pattern (Yanagisawa 2011, 902–904).
A skeletal muscle fiber contracts only when stimulated by a motor neuron. (The generation and propagation of the motor neuron impulse were discussed above.) At rest, regulatory proteins prevent contraction: tropomyosin covers the myosin-binding sites on actin filaments, and the troponin complex controls tropomyosin’s position. When a muscle fiber is stimulated by a nerve, an action potential travels along the sarcolemma and into the fiber via invaginations called T-tubules. This electrical signal causes the sarcoplasmic reticulum (a specialized endoplasmic reticulum) to release Ca²⁺ ions into the sarcoplasm. Calcium binds to troponin, causing troponin and tropomyosin to shift and expose the binding sites on actin. Myosin heads can then attach to actin and perform cross-bridge cycling, leading to contraction. As long as Ca²⁺ remains elevated, contraction continues. When the nerve impulse stops, Ca²⁺ is actively pumped back into the sarcoplasmic reticulum, troponin and tropomyosin re-cover the actin sites, and the muscle fiber relaxes (Campbell 2011, 1152–1153).
A single muscle fiber’s contraction is all-or-nothing, but a whole muscle can contract gradually. We can produce a small or large force with a muscle by recruiting fewer or more motor units. A motor unit consists of a motor neuron and all the muscle fibers it innervates. Each muscle fiber has only one neuromuscular junction, but a motor neuron can connect to many fibers. The contraction of a single muscle cell in response to a single nerve impulse is very brief (tens of milliseconds). To achieve a sustained contraction or greater force, the nervous system can increase the firing frequency of motor neurons. If a second impulse arrives before the muscle fully relaxes from the first, the contractions sum (temporal summation) and tension increases. A high-frequency train of impulses can produce tetanus, a sustained maximal contraction.
Skeletal muscle fibers are classified as fast-twitch (white) or slow-twitch (red) based on their contraction speed and endurance. Fast-twitch fibers contract quickly and powerfully (useful for sprinting or quick, strong movements) but fatigue rapidly. Slow-twitch fibers contract more slowly and with less force, but they resist fatigue and can sustain activity longer (important for endurance activities and maintaining posture). The difference partly comes from calcium handling: slow fibers have a less extensive sarcoplasmic reticulum, so Ca²⁺ reuptake is about five times slower than in fast fibers, resulting in longer contraction duration. Slow-twitch (red) fibers have abundant blood supply, many mitochondria, and high myoglobin content, which supports aerobic metabolism and endurance. Fast-twitch (white) fibers have fewer mitochondria and rely more on anaerobic metabolism. These physiological differences make slow fibers effective for prolonged, low-intensity work and fast fibers for short, high-intensity work (Campbell 2011, 1150–1161).
Now we have a basic understanding of how the nervous system and musculature function. An interesting evolutionary perspective illustrates how complex neural coding underlies even simple motions. For example, to perform the seemingly simple act of extending only the middle finger, the brain doesn’t send a single command to just that finger. Instead, it sends a general command to extend all the fingers of the hand, immediately followed by a command that suppresses extension in all but the middle finger. Evolutionarily, our primate ancestors primarily needed to grasp with the whole hand, so our neural circuits were optimized for gross movements. Finer motor skills (like independent finger movements) evolved later by layering additional inhibitory control onto the basic grip pattern.
This discussion leads us to consider how externally generated electrical currents (like those from a TENS device) can activate muscles — even causing, for example, the middle finger to lift — and thereby potentially relieve pain.
About the operating principle and applications of the TENS device in general
Earlier we noted that ion pumps (like the sodium-potassium pump) are essential for cell function, and that electrolytes (e.g. sodium ions) in the body influence electrical resistance. Life would not be possible without the constant operation of these ion pumps. External electrical currents, such as an electric shock, can disrupt the normal function of ion pumps and excitable cells.
When people refer to an electric shock, they usually mean exposure to mains electricity (for instance, 240 V, 50 Hz AC in Europe). The danger posed by an electric shock depends largely on the current that passes through the body (measured in amperes). For example, if you touch a 240 V live wire with your hand but you are otherwise well insulated (not grounded — say you’re wearing rubber-soled shoes and not touching anything else), you might not feel much because little current flows. However, if your body provides a path to ground (e.g. standing barefoot on a concrete floor or touching a grounded object), that same contact could drive a potentially lethal current through you. Notably, the body responds differently to alternating current (AC) versus direct current (DC), so the type of electrical exposure matters.
Fish and Geddes (2009) reviewed how electrical current is conducted through the human body and provided some illustrative values (using 120 V, 60 Hz AC as an example, which is standard in the USA). Importantly, ~99% of the body’s electrical resistance resides in the skin. Dry intact skin can have a very high resistance (on the order of 100 kΩ or more, especially on thick skin like the palms), whereas the internal body tissues present a much lower resistance (around a few hundred ohms). Any compromise to the skin (cuts, moisture, etc.) drastically lowers resistance. For instance, dry hand skin might exceed 100,000 Ω, while the internal tissue resistance is roughly 300 Ω. An AC voltage above ~500 V can puncture the skin’s protective barrier (causing burns), whereas a high DC voltage might not be felt at all if it doesn’t force enough current through the skin (because it may not break down the skin resistance in the same way).
The human body itself generates small electrical currents — nerve impulses in the brain, heart, muscles, etc. — and it also responds to external electrical stimuli. Electrotherapy takes advantage of this by applying controlled electrical currents to cause physiological changes that can relieve symptoms. Abnormalities in the body’s “electrical” function (manifesting as, say, muscle tightness or pain) can be treated with electrotherapies; TENS is a typical example of low-frequency electrotherapy used for pain relief (Omron 2022).
A TENS device applies a voltage across two electrodes placed on the skin. This voltage drives a current (flow of electrons) through the underlying tissues, stimulating nerves and creating action potentials (nerve impulses). The relationship between voltage (V), current (I), and resistance (R) in this context is governed by Ohm’s law: V = I × R. Voltage is measured in volts (V), current in amperes (A), and resistance in ohms (Ω). Because skin resistance can vary (for instance, moist or well-salted skin conducts better than dry skin), the device’s output adjusts to achieve therapeutic current levels. Typically, effective pain relief with TENS requires on the order of 15–50 mA of current. For example, if the skin-electrode circuit has ~1000 Ω resistance and we want 25 mA of current, we’d need about 25 V (since 0.025 A × 1000 Ω = 25 V). Most battery-powered TENS units can output around 40–60 V, with some reaching up to ~100 V (Quell 2020). The actual current delivered will depend on the skin’s condition and the device’s settings.
TENS therapies are often categorized by their stimulation pattern. Common categories include conventional TENS, acupuncture-like TENS, and intense TENS, distinguished by pulse frequency, intensity, and duration:
Conventional TENS: High-frequency (≈50–100 Hz), low-intensity stimulation (a strong but comfortable tingling sensation), with pulse widths around 50–220 μs.
Acupuncture-like TENS: Low-frequency (≈2–4 Hz), higher intensity (up to the maximum tolerable level, near painful), with longer pulse widths of ~100–400 μs.
Intense TENS: Very high-frequency (~200 Hz), delivered at the highest tolerable intensity, usually for short durations.
The choice of mode can affect the mechanism of pain relief. The key analgesic effects of TENS arise from complex interactions of excitatory and inhibitory neurotransmitters and neuromodulators, which are influenced by stimulation frequency and pattern. TENS primarily targets the nerves that conduct pain (nociceptive fibers), and it can activate both the normal forward-direction signals and antidromic signals (those traveling opposite the usual direction in peripheral nerves). This leads to the release of various neurotransmitters and modulators. Research indicates that TENS analgesia involves multiple systems: it triggers the release of endogenous opioids, serotonin, acetylcholine, norepinephrine, GABA, and it modulates receptor activity (e.g., serotonin 5-HT_2 and 5-HT_3 receptors, δ-opioid receptors in the spinal cord) as well as levels of excitatory amino acids like aspartate and glutamate (Johnson 2007; Johnson 2021).
(The earlier section outlined general neurotransmitter roles; detailed pharmacodynamics, such as receptor affinity, are beyond this scope. In essence, what matters is the balance between neurotransmitter levels and receptor sensitivity — even a small amount of neurotransmitter can be effective if receptors are plentiful, and vice versa.)
Low-frequency TENS (e.g., 2–4 Hz) is reported to be effective for chronic pain, muscle stiffness, and numbness. It promotes the release of endorphins and other pain-relieving substances and improves local blood circulation. The low-frequency pulses cause rhythmic muscle twitches: when muscles relax, fresh blood flows in, and when they contract, waste-laden blood is pushed out. Repeating this cycle enhances perfusion and can disrupt pain signaling to the brain. High-frequency TENS (e.g., 100 Hz or higher), on the other hand, tends to be effective for acute nerve pain, providing faster pain inhibition (Omron 2022).
Current consumer TENS devices are often based on older technological concepts. For example, Omron’s TENS units hold multiple patents and also utilize other companies’ patented technologies. A classic patent in this field (Kolen 1993, via Espacenet 1994) describes a microprocessor-controlled nerve and muscle stimulator. In that design, the device delivers modulated pulses whose strength, frequency, and waveform can be precisely controlled by a microprocessor. This allows the device to store various stimulation programs (for pain relief, reducing swelling, etc.) in memory. The processor can also adjust parameters to minimize discomfort, like the burning sensation that can sometimes occur under electrodes. According to the patent’s background, using electrical impulses for pain relief has been known for a long time (various theories have been proposed since 1965), but the exact mechanisms are still not fully understood. The patent emphasizes that factors such as the waveform shape, pulse repetition frequency, pulse duration, amplitude, and modulation pattern are all crucial for the effectiveness of stimulation.
In general, the appeal of a TENS device is that it can relieve pain without the need for medications, surgery, or invasive procedures. Moreover, TENS has an excellent safety profile with no significant adverse effects reported (Omron 2022).
Studies on TENS treatments
TENS has been studied for decades across a wide range of applications. Johnsson (2021) observed that even after 50 years of research, uncertainty remains about TENS’s clinical efficacy, partly due to variability in study designs and patient populations. Nonetheless, TENS is widely used and recommended in medicine, nursing, and physiotherapy as a non-pharmacological pain management method.
TENS is mainly employed for symptomatic relief in diverse types of pain, regardless of etiology (neuropathic pain, cancer pain, etc.). It has also been explored for non-pain indications such as urinary or fecal retention, constipation, nausea, vomiting, dry mouth (xerostomia), peripheral ischemia and Raynaud’s syndrome, dementia (for certain symptoms), stroke rehabilitation (for sensation/pain issues), edema reduction, wound healing and tissue regeneration, prevention of tissue necrosis, improving sleep, alleviating depression, and even for aiding recovery in coma patients (Johnsson 2021).
Clinical studies and guidelines have reported mixed but often positive findings:
A Current Care guideline (2015) on menstrual pain concluded that TENS can relieve dysmenorrhea (menstrual cramps), with high-frequency TENS being notably effective.
Another Current Care analysis (2012) on knee osteoarthritis found that patients receiving TENS had significantly greater pain relief than controls, based on a meta-analysis (e.g., pooled standardized mean difference ~–0.85 favoring TENS).
Some specific research findings include:
Brien et al. (1984) tested TENS at 80 Hz and 2 Hz on 42 subjects and measured blood β-endorphin levels before, during, and after treatment. They found no significant changes in β-endorphin levels due to TENS, and blocking opioid receptors with naloxone had no effect on pain relief. This suggests TENS’s pain relief in that study was not mediated by systemic β-endorphin release (i.e., it did not raise pain thresholds via circulating endorphins).
Jones et al. (2011) investigated “Acu-TENS” (acupuncture-like TENS) in patients with chronic obstructive pulmonary disease (COPD). In a double-blind trial, they found that compared to placebo, a single session of Acu-TENS improved lung function (FEV₁ increased by ~24% more than placebo), reduced dyspnea (shortness of breath) scores by ~14%, lowered respiratory rate, and significantly increased blood β-endorphin levels in the treatment group. This indicates TENS not only helped breathing parameters but also activated endogenous opioid pathways in COPD patients.
In summary, numerous studies demonstrate various benefits of TENS with minimal disadvantages. At this point, it is fitting to focus on the specific TENS device tested in this project and how it performs in practice.
Omron E2 Elite
Omron Healthcare produces a variety of TENS devices, typically priced around €60–120 depending on features and the retailer. These devices are certified medical devices (meeting CE MDD requirements). The model I examined, the Omron E2 Elite, is an older unit that originally cost about €100. The Omron E2 Elite is an electronic nerve stimulator intended for pain relief. According to the manufacturer, it provides massage-like electrical stimulation to relax muscles or relieve pain (e.g., in the neck and shoulders), and it can reduce physical fatigue and improve local blood circulation in treated areas (Omron 2022).
According to the Omron manual (2022), the E2 Elite has 9 different stimulation programs, selected via specific buttons on the unit:
Quick Relief (blue button): Delivers rapid relief for acute muscle pain using a very high-frequency pulse (~1200 Hz). This mode is suggested for quick pain relief in areas like the lower back or joints.
Zone (yellow button): Offers four region-specific programs tailored for the shoulders, soles of the feet, calves, and lower back.
Massage (green button): Offers four massage simulation modes: Tapping, Kneading, Pressing, and Rubbing, to mimic different massage techniques.
Soft, Repeat, Point, Wide (purple buttons): These four special function buttons modify the automatic programs. Soft reduces intensity, Repeat repeats a stimulation pattern, Point focuses the stimulation in a small area, and Wide spreads the stimulation over a broader area. The device also has “Balance” controls to adjust the relative intensity between the two electrode pads.
Reflecting on research findings, the Omron E2 Elite’s highest frequency program (~1200 Hz) is a relatively high frequency. Based on earlier discussions, high-frequency TENS has been noted as effective for certain types of pain (for example, it was observed to be helpful for menstrual pain in the Current Care 2015 guideline).
Hazards and Warnings
According to the device manual (Omron 2022), the Omron E2 Elite should not be used in certain situations:
Contraindicated Users: Individuals with electronic medical implants (e.g., pacemakers) or other life-supporting devices (e.g., respirators), and those using body-worn medical devices (such as continuous ECG monitors) should not use TENS.
Consult a Doctor First: Users should seek medical advice before using TENS if they have any acute illness, malignant tumor, infectious disease, are pregnant, have heart failure or arrhythmias, have a high fever, abnormal blood pressure, a skin condition or wound at the electrode site, or if they are under medical treatment and feel unwell.
Contraindicated Placement: Do not apply electrodes near the heart, across the front of the neck, on the head, around the mouth, or on irritated/infected skin.
Simultaneous Device Use: Do not use the TENS unit at the same time as other electrical medical devices. Also, ensure the skin is clean and free of creams or lotions before applying electrodes.
These precautions help prevent interference with vital devices and reduce any risks of adverse reactions or improper use (Omron 2022).
Testing the device on my own musculature
After a heavy snowstorm, I shoveled snow for about three hours, which left me with considerable muscle soreness in both arms. This provided an opportunity to test the Omron E2 Elite’s pain-relieving capability. I decided to apply TENS to one arm and use the other arm as an untreated control for comparison.
Test setup
For the experiment, I first attempted to measure what kind of voltages and currents the Omron E2 Elite produces under normal use. Lacking specialized laboratory instruments, I used a Fluke 73III handheld multimeter for basic measurements. (Ideally, one would use an oscilloscope, a capacitance meter, and a high-precision ammeter to fully characterize the device’s output, but those were not available.) The inherent resistance of the multimeter’s test leads is about 0.3 Ω, and the lead wires of the TENS electrodes have a similar small resistance. I used the Omron “Long Life” electrode pads that came with the device, freshly washed to ensure good conductivity. The device was powered with new 2×AAA batteries (3 V supply).
Before using TENS for pain relief, I measured the electrical resistance of my arm’s skin between two electrodes. I placed one electrode near the elbow and the other near the wrist of my left arm (approximately 20 cm apart) where I felt muscle strain. Initially, the multimeter read roughly 200 kΩ between the electrodes; over time (a few seconds) this climbed to about 300 kΩ, presumably as the moisture under the pads diminished or the skin reacted, slightly increasing resistance. Repeated tests suggested that a typical working resistance was around 220 kΩ once things stabilized. (The earlier mention of “100 kΩ vs 1000 Ω” in device specs likely refers to different measurement conditions or misunderstandings, but given the electrode spacing and conditions, a few hundred kilohms of resistance seems reasonable for dry skin contact.)
To better characterize the device output, I selected the “Back” program on the TENS unit, which outputs pulses evenly and at a relatively high intensity. This program gave strong, dense pulses immediately, which was useful because the multimeter’s response is relatively slow (and it could more easily capture a steady reading from a strong continuous pulse pattern). A mode like “Relaxation” on the device uses a more complex, varying pattern that the multimeter would have difficulty tracking.
I then connected a known resistor across the device’s outputs to simulate a stable load. I used a 220 kΩ, 0.5 W resistor (with color code red-red-yellow-brown) in place of my skin. The multimeter confirmed its value at about 224 kΩ (the slight difference accounts for the meter leads and contact). With this resistor as the load, the voltage measured across it was only on the order of a few millivolts, and the multimeter in DC mode showed at most ~0.01 mA of current during the strongest pulses. This low reading indicates that the TENS unit’s pulses are very brief and the multimeter (which averages readings over time) cannot capture the true peak values of these spikes. In other words, an ordinary multimeter isn’t fast enough to measure the short pulses from the TENS; it underreports the actual instantaneous voltage and current of the pulses.
Next, I tried measuring the current directly. I connected the multimeter in series with the electrode output (effectively short-circuiting the device through the meter’s ammeter, which has very low resistance ~1 Ω). On the multimeter’s DC current setting, I again saw around 0.01 mA, similar to before. However, on the AC current setting, the meter fluctuated between approximately 0.01 A and 0.038 A (10–38 mA) when the device intensity was set to 8 out of 10 (which is a very strong setting). The Omron manual indicates the maximum output current is about 40 mA, so seeing peaks up to ~38 mA is consistent with that specification. (According to the manual, 40 mA is the maximum current draw, which also aligns with the battery life expectancy of ~3 months on 2×AAA cells under typical usage.)
Interestingly, if I attempted to measure the device’s output with no load at all (open circuit voltage), the device simply shut off its output. It appears the Omron E2 Elite has a safety feature that detects if the electrodes are not properly attached to a body (i.e., very high impedance) and it stops output to prevent arcing or unexpected shocks. This suggests the device might modulate its output based on load conditions (for instance, increasing voltage if it senses a higher impedance, up to a limit). My rudimentary short-circuit test wasn’t sophisticated enough to determine exactly how the device adjusts to different skin resistances, but the behavior implies some smart sensing.
To relate these findings to theory: if effective pain relief currents are ~15–50 mA (as Quell (2020) mentioned), and if one assumed a ~220 kΩ skin resistance from the simple measurement, naive calculation by Ohm’s law would suggest needing on the order of 3300–11,000 V to drive that current – which is obviously impractical and not what the device does. The resolution to this apparent paradox is that the actual impedance that matters during the brief pulses is much lower (the skin capacitance, tissue impedance, electrode properties, etc., drastically reduce the effective impedance during the fast pulses). It’s likely that either my measurement setup was inadequate for capturing the real dynamics, or Quell’s example assumes a much lower effective tissue resistance (~1000 Ω) which would align with needing 40–120 V, consistent with typical TENS devices. In short, either my measurement was not capturing the real pulse behavior, or the Quell (2020) estimate was simplifying things – or both.
Whatever the case, one might expect that large differences in skin-electrode impedance would affect TENS performance. However, an interesting study by Vance et al. (2015) suggests otherwise. They conducted a blinded trial with 24 subjects receiving TENS at 100 Hz, 100 μs pulse width for 30 minutes, with electrodes placed at optimally effective sites (OSS) and suboptimal sites (SSS). They found significantly different impedance values between optimal and suboptimal placements (the study reported, for example, ~17.7 Ω at the optimal site vs. ~3.5 Ω at the suboptimal site – presumably these values were normalized or measured in a specific way), yet the pain relief reported by subjects was no different. In other words, even though electrode placement altered impedance, it did not change the efficacy of the TENS treatment (Vance, Rakel, Dailey & Sluka 2015). This indicates that, within reasonable ranges, skin impedance might not be a critical factor for TENS effectiveness.
Test results
Ultimately, the primary question was whether the TENS treatment would reduce my muscle pain. I applied the Omron E2 Elite’s “Relaxation” program to my left arm (the test arm) for five minutes. After this TENS session, the pain in my left arm was clearly relieved and did not return for the rest of the afternoon. By contrast, my right arm (which I did not treat with TENS) initially continued to ache as before and only gradually felt better later in the day (presumably as the muscle soreness naturally subsided). This subjective comparison suggests that the TENS treatment provided immediate and noticeable pain relief in the treated arm, whereas the untreated arm followed the slower course of natural recovery.
TENS Device for Pain Relief and Pharmacological Drugs
Given TENS’s potential as a non-pharmacological pain reliever with minimal side effects, one might expect it to be a first-line option for pain management. Johnsson (2021) notes that TENS has shown good outcomes in evidence-based use. Some meta-analyses indicate that in acute pain, TENS can be about 26.5% more effective than placebo (with a confidence range from slightly worse to significantly better, –6% to +51%, reflecting variability in studies). There is also compelling evidence that TENS provides significant pain relief across many chronic pain conditions, though results can vary depending on the condition and study design.
When comparing TENS to drug therapies, it’s informative to consider metrics like NNT/NNH (discussed earlier). For instance, Santana et al. (2016) performed a randomized trial on women in labor and found that using TENS during labor significantly reduced pain and delayed the need for pharmacological analgesia. In that study, the calculated NNT for TENS was approximately 2, meaning that for every two women treated with TENS, one achieved a notable pain reduction (pain rated below 7 on a scale where untreated pain was often above 7). No adverse effects from TENS were observed in that trial.
On the pharmacological side, consider paracetamol (acetaminophen), one of the most common analgesics. A Cochrane review (2022) concluded that paracetamol is essentially ineffective for chronic low back pain: even at a high dosage (4 g per day for 12 weeks), it performed no better than placebo, and yet it poses risks of serious harm (e.g., liver toxicity at high doses). That review did not explicitly provide NNT or NNH, but it underscores that for some conditions, our standard drugs may not be very effective despite being widely used. In terms of acute pain relief, McQuay and Moore (2007) analyzed analgesic trials and found that paracetamol’s NNT for significant pain relief is about 5.4 at a 600 mg dose and improves to 3.7 at a 1000 mg dose. Similarly, aspirin has an NNT of about 4.4 at 600 mg and 4.0 at 1000 mg. Lower NNT values indicate more effective pain relief (an NNT of ~4 means one in four patients benefits, whereas ~3.7 means one in 3.7 patients benefits, etc.).
These figures make one wonder: how many people with back or knee pain routinely take paracetamol (despite its limited efficacy in those cases), and how many women rely on paracetamol for menstrual cramps? It would be very interesting to directly compare TENS to paracetamol in such conditions. Unfortunately, to date no large-scale studies have directly pitted TENS against a common analgesic like paracetamol in a head-to-head trial. In practice, pharmacological pain medication remains the primary approach for most pain conditions, which is reflected in healthcare utilization and spending.
Number Needed to Treat and Number Needed to Harm
When evaluating TENS versus medications, it’s useful to recall the definitions of NNT and NNH (as discussed earlier). Professor Juhani Knuuti (2018) emphasizes that NNT and NNH are straightforward indicators of treatment benefit and risk, respectively. NNT (Number Needed to Treat) is how many patients must receive a treatment for one patient to benefit. An ideal NNT is 1 (meaning every treated patient benefits). NNH (Number Needed to Harm) is how many patients, on average, would experience one adverse outcome. An ideal treatment would have a very high NNH (meaning harms are very rare). It’s important that NNT and NNH are considered in context (what outcome, what condition) and not directly compared to each other unless they pertain to the same outcome.
For example, TENS in labor pain (as per Santana et al. 2016) had an NNT ~2 for delaying pharmacological pain relief, with no observed harms (implying NNH was very high or undefined since no adverse events occurred in that study). On the other hand, paracetamol for certain pains can have a moderate NNT (~3–5 for headache or postoperative pain at typical doses, per McQuay & Moore 2007) but also carries real NNH concerns (e.g., potential liver damage in a small percentage of users, especially if dosed improperly).
Thus, examining acute pain management from both the TENS perspective and the pharmacological perspective is valuable. So far, no direct comparative trials between TENS and standard analgesic drugs have been conducted, but the existing data suggest TENS could be a competitive alternative or adjunct in many scenarios.
Number Needed to Treat and Number Needed to Harm
When comparing TENS with drug therapy, let’s briefly revisit NNT and NNH in a practical context (Knuuti 2018). As defined:
NNT (Number Needed to Treat): The number of patients that need to be treated for one patient to experience the desired benefit.
NNH (Number Needed to Harm): The number of patients that need to be treated for one patient to experience a specific adverse effect.
These metrics are direct measures of a treatment’s effectiveness and safety. For instance, an NNT = 1 means every treated patient benefits (extremely effective), whereas an NNT = 10 means only one in ten benefits (less effective). For harm, an NNH = 1 would be very dangerous (everyone has side effects), while NNH = 1000 indicates side effects are very rare.
It’s also crucial to consider what outcome is being measured. NNT for pain relief and NNT for cure of a disease are not comparable, for example. Similarly, NNH for mild nausea vs NNH for a life-threatening event carry different weight. So these values must be interpreted within their context and not directly compared if they refer to different outcomes.
For TENS vs medications:
TENS example (Labor pain, Santana et al. 2016): NNT ≈ 2 for delaying need for drug analgesia. No significant harms observed (NNH very high or not reached, since no adverse events).
Paracetamol example (Cochrane 2022, McQuay & Moore 2007): For chronic back pain, paracetamol’s benefit seems negligible (effectively NNT is very high or “no better than placebo”). In acute pain, paracetamol’s NNT might be around 4–5 for meaningful relief at common doses. Paracetamol’s NNH for serious harm (like liver toxicity) is thankfully high (serious events are rare when used correctly, but misuse can make NNH quite low).
One striking pharmacological comparison: Cochrane (2022) found paracetamol 4 g/day gave zero benefit over placebo for chronic back pain but certainly can cause harm in some cases — essentially an unfavorable balance (no benefit, some risk). In acute pain, paracetamol does have benefit, but as noted, one might treat 4–5 people for one to get significant relief at standard doses.
For example: TENS vs Paracetamol?
The data above naturally provoke questions. How many people with chronic pain conditions like back pain or knee osteoarthritis are routinely taking paracetamol without significant benefit? How many women take paracetamol for menstrual cramps, when TENS might actually help them more effectively with no systemic side effects? It would be very insightful to see clinical trials directly comparing outcomes between TENS and a common analgesic like paracetamol in such scenarios.
At present, the default approach in medicine is still to start with pharmacological pain relief. This is evident in clinical practice and in tools like symptom checkers, which ask about medication use but often do not mention TENS. It suggests a bias or at least a habit in healthcare to favor drug therapy. The challenge and opportunity moving forward is to integrate TENS as a recognized option, supported by cost-effectiveness analyses and patient education.
About painkillers and costs
Over-the-counter analgesics like paracetamol (acetaminophen) and aspirin are widely used for pain relief, and stronger painkillers (including NSAIDs and opioids) are available by prescription. Even certain antidepressants (in low doses) are used off-label for chronic pain due to their pain threshold-raising effects. Among NSAIDs, ibuprofen is the most commonly used painkiller, followed by paracetamol (though paracetamol is not an NSAID, it is often grouped with them in usage statistics).
To understand the healthcare impact of these drugs, consider paracetamol in detail: Chemically, paracetamol is C_8H_9NO_2 (also written as HOC₆H₄NHCOCH₃). It’s sold under many brand names (Panadol, Tylenol, etc.). Its pharmacological mechanism isn’t completely understood, but it likely involves inhibition of the cyclooxygenase (COX) pathways similar to NSAIDs. One hypothesis is that paracetamol raises the pain threshold by reducing the activity of COX-1 and COX-2 enzymes in the nervous system, thereby decreasing prostaglandin production (prostaglandins enhance pain signals). Paracetamol provides analgesia (pain relief) and is antipyretic (reduces fever) like aspirin, but unlike aspirin it has negligible anti-inflammatory effects and does not affect uric acid levels. Paracetamol’s half-life is about 2.5 hours (for an IV dose of 15 mg/kg), which is why dosing is typically every 4–6 hours. It reduces fever by acting on the hypothalamus in the brain to dissipate heat (through vasodilation and sweating) (PubChem 2021).
In the ATC (Anatomical Therapeutic Chemical) classification, paracetamol’s code is N02BE01 (N = Nervous system, 02 = analgesics, B = other analgesics and antipyretics, E = aniline derivatives (paracetamol group), 01 = specific substance code for paracetamol). Paracetamol is also a component in combination analgesics; for example, codeine-paracetamol combinations have ATC code N02AJ06 (e.g., brand name Panacod) and tramadol-paracetamol combinations have ATC code N02AJ13.
The usage of paracetamol and related drugs can be quantified. According to Finland’s KELA Kelasto database (2020):
There were about 2.12 million reimbursed prescriptions for plain paracetamol (N02BE01) for approximately 855,790 patients, incurring a total cost of about €17.87 million.
Codeine-paracetamol combos (N02AJ06) had about 474,363 prescriptions for 214,812 patients, costing roughly €3.94 million.
Tramadol-paracetamol combos (N02AJ13) had 6,904 prescriptions for 18,557 patients, costing about €202,700.
In total, Finnish healthcare spent on the order of €22 million in 2020 on prescription paracetamol-containing products alone (not counting over-the-counter sales, which would add to this). This reflects substantial usage.
If even a fraction of this medication use could be supplanted by TENS therapy, the economic and health benefits could be significant. For instance, if 20% of those prescription cases (by cost) were replaced with TENS treatments, it could save roughly €4.4 million per year in drug costs. Additionally, reducing medication use would likely reduce drug-related side effects and adverse events. A TENS device is a one-time purchase that can last for years, and it doesn’t introduce pharmaceuticals into the body (thus avoiding issues like drug interactions and environmental contamination via drug excretion).
Medications excreted by patients end up in wastewater and can contribute to environmental pollution. TENS, being a device-based therapy, produces no pharmaceutical waste. The only consumables are batteries (or electricity) and electrode pads, making it a cleaner technology from an environmental standpoint as well.
Development ideas
The Omron E2 Elite operates on 3 V (two AAA batteries) and draws very little current, meaning its power requirements are modest. One development idea is to power such a TENS device via a USB port (5 V supply) and integrate it with a computer or smartphone. This could allow the TENS unit to become a smart accessory: the necessary power could be drawn from the phone or computer, and the control interface could be an app. Through an app or software, it would be easy to update the device’s firmware or add new stimulation modes. If part of a telehealth system, healthcare providers could even remotely adjust the program settings for patients.
Additionally, connecting the TENS device to a smartphone or PC would enable data collection and feedback. Patients could input their pain levels before and after sessions, track usage patterns, and the app could adjust treatment parameters for optimal results. This data could be anonymously uploaded for analysis, giving researchers and clinicians valuable information on TENS efficacy across many users. Overall, such connectivity could personalize treatment and also provide evidence on how well TENS works in real-world use.
Interestingly, products in this direction already exist. The iTENS (2022) is a wearable TENS device controlled through a smartphone app, offering Bluetooth control of the electrode patches on the skin. This confirms that there is a market for app-enabled, wireless TENS units. Professional-grade TENS devices also exist (primarily used by physiotherapists), and many physiotherapists recommend home TENS units to patients. If mainstream healthcare providers (doctors, clinics) are slow to adopt these newer integrated TENS technologies, physiotherapy clinics and consumer markets may drive the change by using and promoting devices like iTENS.
Another observation is that current public digital health services do not yet emphasize TENS. For example, a review of the Finnish ODA project and the OmaOlo online symptom checker (Association of Finnish Local and Regional Authorities 2022; OmaOlo 2022) shows that in the back pain questionnaire, TENS is not mentioned at all. OmaOlo does ask if pain medications provided relief, and likely those responses are collected as statistics. This means all the data being gathered centers on pharmacotherapy outcomes, potentially reinforcing the focus on medications. It would be beneficial for such platforms to start including questions or suggestions about TENS or other non-pharmacological treatments, both to educate users and to gather data on their usage and effectiveness.
Summary
In this report, I examined commonly available TENS devices for pain relief and conducted a hands-on test of the Omron E2 Elite model. The Omron E2 Elite meets medical device CE requirements and is an officially recognized treatment modality for many pain conditions. However, despite more than 50 years of research, strong evidence for certain uses of TENS (like menstrual pain, chronic low back pain, or knee osteoarthritis pain) remains limited to what has been recommended in guidelines (e.g., modest benefits in some cases). This gap in evidence could be due to study design challenges or the inherently modest effect sizes, but it contrasts with the widespread anecdotal and some clinical support TENS has received. It’s also important to critically recognize that modern medicine’s primary approach to pain is pharmacological. This paradigm can make the adoption of treatments like TENS slower, as they may be perceived as contrary to the drug-centered ideology or simply overlooked.
Encouragingly, a search of informal sources suggests that some Finnish hospital districts are already lending TENS devices to patients for home use, hinting at a growing acceptance in mainstream healthcare. At the very least, TENS warrants further examination in the context of the evolving social welfare and healthcare system reforms. There are clear angles to investigate: potential cost savings (fewer drugs, fewer side effects) and the empowerment of patients in self-care (managing pain independently with a device).
The upfront cost of a TENS unit (around €100) might be a hurdle for some individuals. However, for someone who can afford other health-related expenses (such as a gym membership or a hobby), €100 is relatively small compared to the potential benefits of quick pain relief for acute conditions (e.g., muscle strain, tension headaches, etc.). For frequent pain sufferers, a one-time €100 expense could reduce reliance on medications or healthcare visits. Workplaces could even provide TENS devices in break rooms or wellness centers, allowing employees to manage minor musculoskeletal pains on their own and potentially improving productivity. Community health points (such as those run by the Red Cross or local municipalities) could also include TENS therapy as a service for people to try or use on-site.
In conclusion, TENS therapy represents a viable complement or alternative to pharmacological pain management. It is safe, has minimal recurring costs, and fits well with self-care and preventive care models. With further evidence from studies, integration into digital health tools, and increased awareness among both healthcare professionals and the public, TENS could play a larger role in future pain management strategies alongside traditional drug therapies.
Bibliography
Brien, W., Rutan, F., Sanborn, C. & Omer, G. 1984. Effect of transcutaneous electrical nerve stimulation on human blood beta-endorphin levels. Viitattu 14.02.2022. Saatavilla: DOI: 10.1093/ptj/64.9.1367
Campbell. 2011. In Jane B. Reece, Lisa A. Urry, Michael L. Cain, Steven A. Wasserman, Peter V. Minorsky & Robert B. Jackson, Campbell Biology (9th ed.). Pearson, USA.
CERTIPEDIA. 2022. Certipedia – Certificate Database from TÜV Rheinland. Viitattu 25.01.2022. Saatavilla: https://www.certipedia.com/
Cochrane. 2022. Paracetamol: widely used and largely ineffective. Viitattu 01.02.2022. Saatavilla: https://uk.cochrane.org/news/paracetamol-widely-used-and-largely-ineffective
Espacenet. 1994. Microprocessor-based nerve and muscle stimulator for localized application. Viitattu 28.01.2022. Saatavilla: https://worldwide.espacenet.com/patent/search/family/025675962/publication/EP0620025A1?q=pn%3DEP0620025A1
EUDAMED. 2022. EUDAMED – European Database on Medical Devices. Viitattu 25.01.2022. Saatavilla: https://ec.europa.eu/tools/eudamed/#/screen/home
EUR-LEX. 1993. COUNCIL DIRECTIVE 93/42/EEC. Retrieved 2022-01-26. Available: https://eur-lex.europa.eu/legal-content/FI/TXT/HTML/?uri=CELEX:01993L0042-20071011&from=EN
Fimea. 2014. Pharmaceutical Information from FIMEA 2/2014. Retrieved 2022-02-04. Available: https://sic.fimea.fi/documents/721167/862630/26268_2_14_44-45_Vortioksetiini.pdf
Fish, R. ja Geddes, L. 2009. Conduction of Electrical Current to and Through the Human Body: A Review. Viitattu 01.02.2022. Saatavilla: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2763825/
Heino & Vuento. 2018. Basics of Biochemistry and Cell Biology. Sanoma Pro.
Hämäläinen, H. 2017. Lecture: Mental function and mental balance – Neurons, synapses and neurotransmitters. University of Turku, 16.11.2017.
iTENS. 2022. Viitattu 14.02.2022. Saatavilla: https://itens.com/
Kelasto. 2020. Dispensations by ATC class. Retrieved 2022-01-27. Available: https://tilastot.kela.fi/ibmcognos/bi/?perspective=classicviewer&pathRef=.public_folders%2FRaportit%2F91%2BEtuudet%2F9105%2BL%C3%A4%C3%A4kkeet%2FSairausvakuutuksesta%2Bkorvattavat%2Bl%C3%A4%C3%A4ketoimitukset%2B%25289105RS001%252529&id=i1465D29B12B943909136FA98D42354D8&objRef=i1465D29B12B943909136FA98D42354D8&action=run&format=HTML
Association of Finnish Local and Regional Authorities. 2022. Through the ODA project, social and health care services are taking a leap into the digital age. Retrieved 2022-02-14. Available: https://www.kuntaliitto.fi/sosiaali-ja-terveysasiat/oda-projektin-kautta-sosiaali-ja-terveyspalvelut-loikkaavat-digiaikaan
Current Care. 2014. TENS in chronic low back pain. Retrieved 2022-02-14. Available: https://www.kaypahoito.fi/nak08252
Current Care-b. 2019. Physiotherapy for sensory disorders and pain in MS disease. Retrieved 2022-02-14. Available: https://www.kaypahoito.fi/nak05367
Current Care-c. 2015. Transcutaneous electrical nerve stimulation (TENS) in the treatment of menstrual pain. Retrieved 2022-02-14. Available: https://www.kaypahoito.fi/nak08495
Current Care-d. 2012. The effect of TENS electric current on pain caused by knee osteoarthritis. Retrieved 2022-02-14. Available: https://www.kaypahoito.fi/nak07846
Linnavuori, K. 2015. A new EU regulation on medical devices. Retrieved 2022-01-25. Available: https://www.fimea.fi/documents/160140/765540/28338_Linnavuori_ATMP_2015-02-04_2_.pdf
McQuay, H. & Moore, R. 2007. Dose–response in direct comparisons of different doses of aspirin, ibuprofen and paracetamol (acetaminophen) in analgesic studies. Viitattu 01.02.2022. Saatavilla: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2000740/
MDR. 2022. Medical Device Regulation – Conformity assessment based on a quality management system and on assessment of technical documentation. Viitattu 27.01.2022. Saatavilla: https://www.medical-device-regulation.eu/2019/08/14/annex-ix/
MDR-b. 2022. Medical Device Regulation – Classification rules. Viitattu 27.01.2022. Saatavilla: https://www.medical-device-regulation.eu/2019/08/08/annex-viii/
Johnsson, M. 2007. Transcutaneous Electrical Nerve Stimulation: Mechanisms, Clinical Application and Evidence. Viitattu 27.01.2022. Saatavilla: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4589923/
Johnsson, M. 2021. Resolving Long-Standing Uncertainty about the Clinical Efficacy of Transcutaneous Electrical Nerve Stimulation (TENS) to Relieve Pain: A Comprehensive Review of Factors Influencing Outcome. Viitattu 01.02.2022. Saatavilla: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8070828/
Johnson, M., Paley, C., Howe, T. & Sluka, K. 2015. Transcutaneous electrical nerve stimulation for acute pain (Cochrane Review). Viitattu 14.02.2022. Saatavilla: DOI: 10.1002/14651858.CD006142.pub3
Jones, A., Ngai, S., Hui-Chan, C. & Yu, H. 2011. Acute Effects of Acu-TENS on FEV1 and Blood β-endorphin Level in Chronic Obstructive Pulmonary Disease. Viitattu 14.02.2022. Saatavilla: https://pubmed.ncbi.nlm.nih.gov/22314671/
Knuuti, J. 2018. Overdiagnosis and overtreatment: Number needed to treat. From the blog Health & Science. Retrieved 2022-02-01. Available: https://blogit.ts.fi/terveys-tiede/ylidiagnostiikka-ja-ylihoito-osa-2/
Loh, H., Tseng, L., Wei, E. & Li, C. 1976. Beta-endorphin is a potent analgesic agent. Viitattu 14.02.2022. Saatavilla: doi: 10.1073/pnas.73.8.2895
OmaOlo. 2022. Low back pain or injury (online symptom assessment). Retrieved 2022-02-14. Available: https://www.omaolo.fi/palvelut/oirearviot/2
Omron. 2022. Electronic Nerve Stimulator E2 Elite (HV-F127-E) instruction manual. (Appendix 2 of this report.)
Paavilainen, P. 2016. A functioning brain. In Paavilainen, P. Toimiaivot – Basics of Cognitive Neuroscience. 1st ed. Edita, pp. 1–368.
Penttonen, M. 2017. Lecture: Nervous System Development and Plasticity. University of Jyväskylä, 21.11.2017.
PubChem. 2021. Acetaminophen. Retrieved 2022-01-31. Available: https://pubchem.ncbi.nlm.nih.gov/compound/1983
Quell. 2020. TENS Under the Hood: Maximum Voltage. Retrieved 2022-01-27. Available: https://www.quellrelief.com/blog/tens-under-the-hood-maximum-voltage/
Santana, L., Gallo, R., Ferreira, C., Duarte, G., Quintana, S. & Marcolin, A. 2016. Transcutaneous electrical nerve stimulation (TENS) reduces pain and postpones the need for pharmacological analgesia during labour: a randomised trial. Viitattu 14.02.2022. Saatavilla: https://www.sciencedirect.com/science/article/pii/S1836955315001289
Vance, C., Rakel, B., Dailey, D. & Sluka, K. 2015. Skin impedance is not a factor in transcutaneous electrical nerve stimulation effectiveness. Viitattu 01.02.2022. Saatavilla: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4547643/
Yanagisawa, M. 2011. In Jane B. Reece, Lisa A. Urry, Michael L. Cain, Steven A. Wasserman, Peter V. Minorsky & Robert B. Jackson, Campbell Biology (9th ed.). Pearson, USA.
Wu, L., Weng, P., Chen, C., Huang, Y., Tsuang, Y. & Chiang, C. 2018. Literature Review and Meta-Analysis of Transcutaneous Electrical Nerve Stimulation in Treating Chronic Back Pain. Viitattu 14.02.2022. Saatavilla: DOI: 10.1097/AAP.0000000000000740