Orthopedics · Part One · How Pain Really Works
Lesson 04 / 44
The Biopsychosocial Model: How Biology, Psychology and Social Context Produce Pain Together
Pain is not a readout of tissue damage. It is a protective output the nervous system assembles from biology, psychology, and social context together.
The biopsychosocial model holds that pain arises from biological, psychological and social processes acting together, rather than from tissue damage alone. George Engel proposed it in Science in 1977 as a replacement for the biomedical model. Among 516 hamstring injuries in professional football, 13 percent showed no abnormality on MRI at all. The Unified Model of Tone supplies the variable the three domains are readings of, which is tone.
Hamstring injuries with no abnormality visible on MRI
13 percent of 516
Strongest single predictor of persistent disabling back pain
nonorganic signs, likelihood ratio 3.0
Amputees reporting phantom pain in the previous four weeks
79.9 percent of 914
Stratified care against current best practice at 12 months
1.06 points, at lower cost
Biopsychosocial and psychosomatic
Biopsychosocial describes a method of assessment. Biology, psychology and social circumstance are all measured, because all three move the output. Psychosomatic describes a supposed origin, a physical symptom produced by the mind. The first assigns no cause and rules nothing out. Serious disease is screened for with the same rigor either way.
How a belief reaches a tissue
Threat appraisal sets the level of protection the nervous system runs. A body part believed to be fragile gets guarded, loaded less, and reported on by a thinner afferent stream. A system reading less information raises protection further and produces pain at a lower threshold. The belief reaches the tissue through behavior, and the tissue reports back.
01Three domains, one patient
The biopsychosocial model replaced the biomedical model in 1977
Pain emerges from biological, psychological and social factors interacting at once, never from damaged tissue alone. That answers the real question a patient carries into the room. If the scan looks bad, why does the pain not match, and what does that mean going forward?
George Engel put the argument in Science in 1977. The dominant model of disease was biomedical, and it left no room for the social, psychological and behavioral dimensions of illness (Engel 1977).
The framework became the mainstream one. A review in Psychological Bulletin called it the most heuristic approach to chronic pain then available (Gatchel 2007). Naming the three domains is where the standard account stops.
What each layer contains
The biological layer includes discs, muscles, tendons, joints, and the nociceptors that report mechanical, thermal and chemical stress. The psychological layer includes beliefs, expectations, fear, catastrophizing, and the memory of prior injuries. The social layer includes what parents modeled, what previous clinicians said, what an imaging report implied, and what the surrounding culture treats as dangerous.
A skilled clinician reads all three, because pain is the nervous system’s opinion about how much protection the body needs, not a direct measure of harm.
02Findings
What the research shows
From a founding paper in Science, an injury cohort, an amputee survey, three systematic reviews and two stratified-care trials.
03Treating the picture
Reading the scan as the diagnosis turns ordinary findings into damage
A purely mechanistic model assumes that pain equals structural damage, and that assumption quietly harms people. It traces back to Descartes and his seventeenth century specificity theory, which pictured pain as a straight line from damaged tissue to a specific pathway to a fixed sensation. A Short History of Pain follows that lineage.
The mismatch runs both ways. Many people carry significant degeneration on MRI with little or no pain, and many others report severe pain with almost no tissue damage.
What the diagnosed injuries showed
Ekstrand and colleagues followed 23 European professional teams from 2007 to 2011 and recorded 516 hamstring injuries, of which 58 percent went to MRI (Ekstrand 2012). Thirteen percent were graded 0, meaning nothing visible at all. Seventy percent were grade 0 or 1, with no sign of fiber disruption anywhere in the muscle and tendon complex.
Those two grades caused 56 percent of the total lay-off, 2141 days of 3830. The scan graded severity well and settled the existence of a problem poorly.
When a clinician treats the picture instead of the person, imaging findings become diagnoses, normal age related changes get framed as damage, and the patient inherits fear. That inheritance is measurable. What a clinician believes about back pain tracks what that clinician’s patients believe (Darlow 2012).
That fear is not harmless. It changes movement, recruitment, and behavior, and it can build the very pain it claims to explain. Degenerative changes appear on the scans of people with no symptoms at every adult age, as Findings in People Without Pain sets out.
04Pain without a limb
Pain in an absent limb is common, severe, and graded by mood
Phantom limb pain settles the argument that pain can exist without tissue. People feel vivid, sometimes severe pain in a limb that is no longer present, which means the tissue cannot be the source.
Ephraim and colleagues interviewed 914 people with limb loss by telephone between 1998 and 2000 (Ephraim 2005). Ninety-five percent reported amputation-related pain in the previous four weeks. Phantom pain led at 79.9 percent, with residual limb pain at 67.7 percent and back pain at 62.3 percent.
Across every pain site, depressive symptoms predicted both the intensity and the bothersomeness of the pain. A psychological measure graded the severity of pain in a limb that no longer existed.
Where the experience is assembled
The experience is generated by the brain, within a distributed network Ronald Melzack named the pain neuromatrix. It spans limbic centers for emotion, autonomic centers for the visceral response, sensory cortex for location and cognitive cortex for evaluation.
The same model explains why rubbing a sore spot helps. Gate control theory, published by Melzack and Wall in 1965, showed that other sensory input and descending signals from the brain can close a spinal gate and turn pain down.
The brain runs its own pharmacy as well, releasing natural opioids from structures such as the periaqueductal gray. Those descending pathways quiet traffic at the first relay in the cord, as Pain Is Not Tissue Damage sets out. Pain is actively regulated, which is precisely why it responds to more than a scalpel.
05Yellow flags
Psychosocial findings predict persistent disabling back pain better than structural ones do
Yellow flags are the psychosocial prognostic factors for disability developing after musculoskeletal pain begins (Nicholas 2011). Later formulations narrowed the term to psychological risk factors, separating them from social and environmental variables.
Chou and Shekelle pooled 20 studies and 10,842 patients with fewer than eight weeks of low back pain (Chou 2010). Nonorganic signs carried a median likelihood ratio of 3.0 for persistent disabling pain at one year, maladaptive coping 2.5, psychiatric comorbidity 2.2 and low general health status 1.8.
The structural and demographic variables ranked below all of them. Radiculopathy, baseline pain and work environment sat near 1.5. A history of prior episodes sat near 1.0, alongside age and sex. A review of 18 prospective cohorts agreed, implicating distress, depressive mood and somatization (Pincus 2002).
The fear cycle
The fear-avoidance cycle is how a biopsychosocial problem becomes a chronic one. It begins with a belief, often planted by a careless phrase such as you have a weak back, that a body part is fragile and that hurt means harm. The person guards the area and avoids loading it.
Reduced use leads to deconditioning, and deconditioning means lost strength, poorer coordination, and less sensory information reaching the brain. The nervous system reads a weaker, quieter tissue and concludes it must protect it more, so it lowers its threshold and produces pain sooner. Protection rises and capacity falls.
Each turn of the loop confirms the original fear. Low fear avoidance was one of the two most useful predictors of recovery at one year, at a likelihood ratio of 0.39 (Chou 2010). Naming this cycle is not blaming the patient. It is identifying the exact mechanism that conservative, confidence building care is designed to reverse, and Hurt Is Not Harm sets out the graded work.
06The social layer
Education, income and the consultation itself move back pain outcomes
The social domain has the largest evidence base and the least clinical attention. A review of 41 studies covering 2,161,617 adults across 17 countries mapped 166 relationships between social factors and chronic low back pain (Karran 2020). Educational attainment and socioeconomic status carried the strongest evidence.
The consultation belongs to the same layer. Clinicians with a biomedical orientation or elevated fear avoidance beliefs more often advised patients to limit work and physical activity, and less often followed treatment guidelines (Darlow 2012). What a clinician says becomes part of the patient’s social input.
The reviewers reported their weak points plainly. Twenty-four percent of the relationships carried a high risk of bias from confounding and 19 percent from missing data. The direction held across 17 countries anyway.
07STarT Back stratification
Sorting patients by psychosocial risk changed outcomes in England and not in the United States
The biopsychosocial model became measurable when it was turned into a screening tool. The STarT Back tool holds 9 items, and its last five form a psychosocial subscale covering bothersomeness, catastrophizing, fear, anxiety and depression (Hill 2008).
Scores of 0 to 3 mark low risk, 4 or 5 on the psychosocial subscale marks high risk, and everyone else is medium risk. The tool was developed in 131 patients and validated in an independent sample of 500.
The trial that followed matched each risk group to a different pathway. Hill and colleagues randomized 851 adults from ten general practices in England, 568 to stratified care and 283 to control (Hill 2011). Disability improved by 4.7 points against 3.0 at four months, a difference of 1.81 points, and by 4.3 against 3.3 at twelve months, a difference of 1.06.
What happened in the United States
The same strategy was implemented in six primary care clinics in the United States and produced nothing. Three clinics were randomized to training and three to control (Cherkin 2018). Clinicians attended six didactic sessions, physical therapists received five days of training, and the tool sat inside the electronic health record.
Clinicians used the tool for about half of their patients. They did not change the treatments they recommended. The intervention had no significant effect on back-related function, pain severity or health care use at the six month endpoint.
The two trials ran the same instrument on the same kind of patients. In England the sorting changed which treatment each group received, and in the United States it did not.
08Claims removed from this page
Three claims from the earlier version were removed
A gold pull-quote attributed to Dr. Jason Dulberg came off the page, because its wording could not be matched to any recorded source. The idea inside it survives as our model’s claim: the same load meets a different state and produces a different outcome. The figure that about 20 percent of diagnosed hamstring injuries show no visible tissue damage came off as well. The published series puts grade 0 at 13 percent of 516 injuries (Ekstrand 2012), and the larger finding is the 70 percent with no fiber disruption.
The earlier text credited the biopsychosocial model to the work of researchers including Melzack, Wall, Lorimer Moseley and David Butler. Those four built the modern science of pain. The framework itself was proposed by George Engel in 1977, and Engel was arguing about medicine as a whole rather than about pain (Engel 1977).
09Why this reasoning protects
Three-domain reasoning keeps an incidental finding from becoming an operation
Reasoning biopsychosocially is what protects patients from over imaging and over treatment. When a clinician understands that structural findings and pain are poorly correlated, an incidental disc bulge or a degenerative change stops being an automatic call for injections or surgery. The scan becomes one piece of information interpreted alongside how the symptoms actually behave, never a verdict on its own.
This is the intelligence behind conservative first. Restoring movement, rebuilding capacity, calming a sensitized system, and replacing fear with accurate understanding address every layer the model describes, and they do so without closing any door.
If conservative care resolves the problem, the person is spared an irreversible intervention. If it does not, the serious pathologies have been screened for and ruled out with care before anything invasive is on the table.
The economics ran the same way. Stratified care cost 240.01 pounds per patient against 274.40, while gaining 0.039 quality-adjusted life years (Hill 2011). Careful reasoning is not the slow path. It is the safe and intelligent one.
10The model on the three domains
What the Unified Model of Tone claims about the biopsychosocial model
Everything above is established science, including the trial that came back null. What follows is our model’s reading, stated as ours.
The biopsychosocial model names three domains and leaves them unjoined. It says biology, psychology and social context all matter, and it never says what they are all readings of. That gap is why it is so often applied as three separate checklists.
Our model supplies the missing variable. Tone is the integrated organization through which the body’s many interacting processes relate to one another at a given moment. It gathers mechanical tension, neural excitability, autonomic regulation, sensory gain, prediction and behavioral readiness. Biology, psychology and social context are three windows onto that one organization.
The chord is not the notes. The biopsychosocial model names the notes well, and a patient walks in with the chord. A disc finding, a belief about fragility and a job that cannot be missed are three inputs to one regulated variable, and the variable is what hurts.
That gives one equation rather than two definitions. Health is regulated tone and disease is dysregulated tone, and the three domains are the channels through which one organization reports itself.
The prediction
From that follows a claim the biopsychosocial literature does not make. Our model predicts that four measures drawn from all three domains, recorded at once in the same person, load on one common factor rather than varying independently.
The four are pressure pain threshold measured away from the painful region, the STarT Back psychosocial subscale score, resting heart rate variability, and time to return to baseline after a standardized load test. No biopsychosocial account requires a questionnaire score to track heart rate variability. Our model requires it.
The same reading explains the null. Six clinics sorted their patients correctly and delivered the same treatments as before (Cherkin 2018). An input interacting with a tone creates an outcome, and a category is not an input.
This is a claim about how pain is organized rather than a claim about what treatment does. If pressure pain threshold, the STarT Back psychosocial subscale, resting heart rate variability and time to return to baseline after a load test are shown to move together, the unification claim is confirmed.
11The tone reading
How the biopsychosocial model expresses tone
Every topic in this library expresses all of tone. In the biopsychosocial model three aspects carry the signature, because measures from the three domains keep predicting one another.
Coupling
Mood and pain move together rather than in parallel. Among 914 people with limb loss, depressive symptoms predicted pain intensity at every site, including limbs no longer present.
Prediction
What a person expects of recovery sets the level of protection. Clinician beliefs about back pain tracked patient beliefs across 17 studies from eight countries.
Input quality
A risk score is information, not treatment. Six United States clinics ran a validated tool on about half their patients and recommended exactly the same treatments.
The remaining foundations run through the biopsychosocial model as well. Constraint: guarding narrows the movements available, and the narrowing becomes its own source of nociception. Gain: the amplifier is set by the whole situation, which is why the same finding hurts differently at work and at home. Set point: the resting level of protection decides how much stimulus it takes to hurt. Time course: psychosocial risk measured in the first eight weeks predicts disability a year later. Load: what a job demands of a body belongs to the biological layer and the social layer at once. Oscillation: pain that follows a daily rhythm is reporting the regulator rather than the tissue. These are readings of one organization rather than three separate systems, which is the core claim of the Unified Model of Tone.
12Across the library
How this page relates to the rest of the library
The three domains reappear across this section under other names.
The output thesis this framework rests on, with the emergency clinic and imaging data behind it.
Descartes, gate control theory and the neuromatrix, the lineage that produced the modern view.
The mechanism inside the biological layer when protection stays high after the tissue settles.
The graded exposure work that reverses the fear cycle described on this page.
The age-stratified prevalence of spinal findings in people with no symptoms at all.
Why the least invasive input that can carry the message belongs first.
Why the same injury in two people produces two courses, read as one regulatory state.
13Frequently asked
Questions patients ask about the biopsychosocial model
What is the biopsychosocial model of pain?
It is the understanding that pain arises from biological, psychological and social factors together, including tissue, beliefs, fear, stress and context, rather than from tissue damage alone. George Engel proposed it in Science in 1977 as a replacement for the biomedical model, which he argued left no room for the social, psychological and behavioral dimensions of illness. He offered it as a blueprint for research and a design for action in health care. A review in Psychological Bulletin thirty years later called it the most heuristic approach to chronic pain then available.
Do stress and beliefs really affect back pain?
Yes, and they outrank most structural findings as predictors. Across 20 studies and 10,842 patients with recent low back pain, nonorganic signs carried a median likelihood ratio of 3.0 for persistent disabling pain at one year and maladaptive coping carried 2.5. Psychiatric comorbidity carried 2.2 and low general health status 1.8. Radiculopathy sat near 1.5. A history of prior episodes, age and sex sat near 1.0, which means they carried no useful information about who would still be disabled a year later.
Why does my clinician ask about sleep and stress for back pain?
Because the questions that predict a poor outcome are mostly not about the tissue. The validated STarT Back screening tool holds nine items, and its psychosocial subscale covers bothersomeness, catastrophizing, fear, anxiety and depression. A score of 4 or 5 on that subscale marks high risk of persistent disabling pain, and that classification decides which pathway a person is offered. Asking about daily life is how a clinician reaches that decision, rather than an inquiry into character or an alternative to examining the spine.
Does the biopsychosocial model mean my pain is psychological?
No. Biopsychosocial describes how an assessment is conducted, and psychosomatic describes a supposed origin in the mind. The first assigns no cause and rules nothing out, and serious disease is screened for with the same rigor either way. The clearest example runs the other direction entirely. Among 914 people with limb loss, 79.9 percent reported phantom pain in the previous four weeks, and depressive symptoms predicted its intensity in a limb that no longer existed. Psychology grades pain that has no tissue to come from.
Can a scan be normal when the injury is real?
Routinely. Across 516 hamstring injuries in 23 European professional football teams, 13 percent were graded 0 on MRI, meaning nothing visible at all, and 70 percent showed no sign of fiber disruption. Those two grades accounted for 56 percent of all lay-off days, 2141 days of 3830. A clean scan did not mean a short absence. Time lost still ran 8 days at grade 0 and 73 days at grade 3, so the grading predicted duration while missing existence.
What are yellow flags?
Yellow flags are the psychosocial prognostic factors for disability developing after musculoskeletal pain begins, and later formulations narrowed the term to psychological risk factors specifically. They include catastrophizing, fear of movement, low expectation of recovery, distress and depressed mood. Screening for them early was meant to trigger secondary prevention before disability sets in. Targeting them produces more consistent results when they are present at high levels than either ignoring them or giving the same package to everyone regardless of risk.
What does the Unified Model of Tone say about the biopsychosocial model?
The three domains are readings of one variable rather than three separate causes. Tone is the integrated organization through which the body’s interacting processes relate at a given moment, and the chord is not the notes. Health is regulated tone and disease is dysregulated tone, one equation rather than two definitions. Our model predicts that pressure pain threshold, the STarT Back psychosocial subscale, resting heart rate variability and recovery time after a load test share one underlying factor. That concerns how pain is organized rather than what any treatment does.
14The sources
References
12 primary sources, each linked to its record. Figures quoted on this page were checked against the published abstract.
Related evidence