One of the most common things I hear in clinic is some version of this: "I've had every scan done. Nothing shows up. But the pain is definitely there." It's frustrating — and entirely consistent with what the research on chronic pain has been telling us for over two decades.
Pain that persists beyond the normal healing window rarely has a single structural cause. It is almost always the result of multiple converging systems — tissue load, nervous system state, sleep quality, and psychological stress — all amplifying each other simultaneously. Understanding why this happens is not just reassuring; it is clinically useful. It changes what treatment is actually targeting.
The Old Model — and Why It Falls Short
For most of the twentieth century, pain was treated as a direct read-out of tissue damage: more damage, more pain; fix the damage, remove the pain. This is sometimes called the biomedical model. It works adequately for acute injuries, where the relationship between tissue state and pain signal is relatively tight.
It breaks down completely for persistent pain. Studies consistently show that many people with severe structural findings on imaging report no pain, while others with minimal structural changes experience debilitating, widespread pain. In a landmark review, Jensen and colleagues (1994) found that 52% of asymptomatic adults had disc bulges on MRI, and 38% had disc abnormalities — none of which produced any pain at all.1 More recently, Brinjikji et al. (2015) pooled data from 3,110 asymptomatic individuals and found that disc degeneration, disc bulge, and disc protrusion are present in the majority of adults over 50 — as an ordinary feature of ageing, not a predictor of pain.2
Structure alone does not explain the pain experience. Something else is going on.
The Biopsychosocial Model
The framework that replaced the biomedical model — and that now underpins musculoskeletal research and clinical guidelines globally — is the biopsychosocial model, originally proposed by Engel (1977) and applied comprehensively to pain by Waddell (1987) and Turk (1996).3,4 It recognises that pain is not simply a signal from damaged tissue, but an output produced by the brain, based on all available inputs — biological, psychological, and social.
In practical terms, this means four overlapping systems typically contribute to a persistent pain presentation:
Structural load, postural strain, or local inflammation generating nociceptive signals from the periphery. A real, measurable contribution — but rarely the whole story in persistent pain.
Amplification of pain signals within the spinal cord and brain. The nervous system itself becomes hyperresponsive — producing pain in excess of what peripheral input alone would generate.
Chronic activation of the hypothalamic–pituitary–adrenal (HPA) axis. Elevated cortisol and pro-inflammatory cytokines measurably lower the pain threshold at the spinal cord.
Impaired descending inhibition and elevated inflammatory markers resulting from poor sleep. Shown to be a predictor — not merely a consequence — of pain severity.
Central Sensitisation: When the Volume Gets Turned Up
Central sensitisation is the most important concept in understanding why persistent pain is so difficult to resolve with tissue-level treatment alone. The term was coined by Clifford Woolf and colleagues in the 1980s and has been refined extensively since. In Woolf's landmark 2011 review in Pain, central sensitisation is defined as "an amplification of neural signalling within the CNS that elicits pain hypersensitivity" — occurring in the absence of ongoing tissue damage.5
In practice, this means the spinal cord and brain become hyperexcitable. Signals that would normally not produce pain are experienced as painful (allodynia). Signals that would produce mild pain produce severe pain (hyperalgesia). The nervous system has, in effect, recalibrated its gain upward — not because something is broken, but because sustained input over months or years has trained it to be on high alert.
Key finding: Central sensitisation does not require ongoing tissue damage to be maintained. Once established, it can persist independently — which is why treating only the structural component often produces incomplete or short-lived relief. (Woolf, 2011; Nijs et al., 2021)5,6
A validated screening tool — the Central Sensitisation Inventory (Mayer et al., 2012) — is now used in clinical research and allied health settings to identify the degree of sensitisation present, and to guide treatment accordingly.7
Sleep: Not a Byproduct — a Driver
Most people understand intuitively that pain makes sleep difficult. What is less well known is that the relationship runs just as strongly in the other direction: poor sleep directly amplifies pain sensitivity, independent of how much pain you currently have.
In a comprehensive meta-analysis across 78 published studies, Finan, Goodin, and Smith (2013) in the Journal of Pain found that sleep disturbance was a stronger prospective predictor of pain than pain was of sleep disturbance — and that the association was bidirectional and self-reinforcing.8 The mechanism involves both reduced descending pain inhibition (the brain's natural pain-off switch, which is consolidated during deep sleep) and elevated inflammatory markers including interleukin-6 and TNF-alpha, both of which sensitise peripheral nociceptors.
A separate experimental study by Smith et al. (2007) demonstrated that even a single night of sleep restriction in healthy individuals significantly increased pain sensitivity the following day — measurable using standardised pressure-pain threshold testing.9 This has direct clinical relevance: a client who presents having slept poorly for months is not just tired. Their baseline pain sensitivity will be measurably elevated regardless of their tissue state.
Stress, Cortisol, and the Pain Threshold
Psychological stress is not merely a psychological phenomenon. Its downstream biological effects are well characterised and include direct modulation of pain processing at the spinal cord level.
Under chronic stress, the hypothalamic–pituitary–adrenal (HPA) axis maintains elevated glucocorticoid (cortisol) output. Cortisol at normal levels is anti-inflammatory; under chronic elevation, it paradoxically contributes to a pro-inflammatory state through glucocorticoid receptor desensitisation and upregulation of inflammatory cytokines including IL-1β, IL-6, and TNF-α.10 These cytokines directly sensitise peripheral nociceptors — lowering the threshold at which pain signals are generated — and contribute to maintaining the central sensitisation state described above.
Loggia and colleagues (2008) demonstrated that experimental psychological stress reliably reduces pressure-pain thresholds in healthy participants — a direct measurable effect of HPA activation on nociception.11 Importantly, this effect is not mediated by catastrophising or attention alone; it is present even under conditions designed to control for psychological confounds.
What this means clinically: When a client's pain is worse during high-stress periods — even without any change in activity or loading — that is not coincidence. It is a measurable biological effect of cortisol on the pain-processing system. Addressing stress load is part of addressing the pain itself.
The Cycle That Keeps It Going
These four systems — tissue input, central sensitisation, stress, and sleep — do not act in isolation. They form a mutually reinforcing cycle. Ongoing pain elevates stress hormones and disrupts sleep. Poor sleep amplifies pain sensitivity and impairs recovery. Elevated stress hormones lower the pain threshold and drive sensitisation. Each factor worsens the others, and the whole system moves further from its normal calibration.
Keeps the central nervous system on high alert; reinforces sensitisation
Reduces descending inhibition; elevates inflammatory markers
Sustained cortisol output sensitises peripheral nociceptors
Pain is experienced earlier and more intensely; cycle repeats
Breaking any link in this cycle — improving sleep, reducing stress load, reducing peripheral nociceptive input through manual therapy, or reducing sensitisation through pain education — has a dampening effect on the whole system.
Pain Education as a Treatment in Itself
One of the most robust findings in modern pain science is that simply understanding the neuroscience of pain produces measurable clinical benefit. This isn't a soft observation — it has been demonstrated repeatedly in randomised controlled trials.
Louw and colleagues (2011) conducted a systematic review of neuroscience pain education (NPE) across multiple RCTs and found that NPE significantly reduced pain intensity, disability, catastrophising, and movement avoidance — independent of any hands-on treatment.12 The proposed mechanism is that accurate knowledge of central sensitisation reduces threat appraisal — the brain's assessment that the pain signal represents genuine danger. Lower threat appraisal reduces the gain on the sensitised system.
Moseley and Butler (2015), in their fifteen-year review of pain education research in the Journal of Pain, concluded that "understanding pain biology is associated with reduced catastrophising, increased physical performance, and changed brain activity patterns consistent with reduced central sensitisation".13 Education does not merely help patients cope with pain — it begins to change the neurological mechanisms producing it.
What This Means for Your Treatment
None of this makes pain "all in your head." Central sensitisation is a measurable neurological phenomenon. Sleep-driven pain amplification is biochemically documented. HPA-mediated nociceptive sensitisation has been demonstrated in controlled trials on healthy subjects. These are real, physical mechanisms — they happen in the body, not in spite of the body.
What it does mean is that treating only the tissue — only the tight muscle, only the restricted joint — addresses one input into a multi-input system. It often helps. But for persistent pain, getting a lasting result typically requires addressing the sleep component, the stress load, the fear-avoidance patterns, and the peripheral tissue load simultaneously. That is what a biopsychosocial-informed assessment is designed to identify, and why treatment approaches grounded in this framework consistently outperform single-modality treatment in clinical trials for chronic pain.4,6
Your nervous system is not damaged. It is trained — and like any trained system, it can be retrained. That process usually starts with understanding what you're actually dealing with.