Orthopedics · Part One · How Pain Really Works
Lesson 05 / 44
Hurt Is Not Harm: Graded Activity, Fear of Movement, and How a Clinician Judges a Flare
One of the most liberating ideas in modern pain science is also one of the simplest. Pain is a protective system, not a damage meter, and understanding that difference can begin to change a person’s experience before a single hands-on technique is applied.
Hurt is not harm is the clinical rule that pain during movement does not by itself indicate tissue damage. Across seven trials and 385 participants, exercise protocols that allowed pain produced slightly better short-term pain outcomes than pain-free protocols. Graded activity and graded exposure apply the rule in practice, progressing load on a quota rather than on how a day feels. The Unified Model of Tone judges the response by function and capacity.
Painful against pain-free exercise protocols, short-term pain
standardized mean difference 0.27
Ordinary activity against two days of bed rest
faster recovery at three and twelve weeks
Tampa Scale for Kinesiophobia, items kept in the short form
11 of the original 17
Graded activity against control, sick leave in the second follow-up year
12.1 weeks against 19.6
Graded activity and graded exposure
Graded activity sets a starting quota of work a person can manage on a poor day, then raises it on a fixed schedule. Progression follows the schedule and not the symptom. Graded exposure ranks activities by how harmful the person expects them to be, then works up that hierarchy. One targets tolerance. The other targets the expectation.
The fear-avoidance cycle
Pain-related fear splits behavior two ways. Confrontation puts a person back in contact with the feared movement, and the fear falls across repetitions. Avoidance removes that correction, so the expectation of harm stands and spreads to more movements. Hypervigilance, guarded movement and withdrawal from activity follow, and disability grows while the tissue stays the same.
01Loading into pain
Movement that hurts can still be safe, and the trials tested it directly
Pain during movement does not by itself mean tissue is being damaged. Nine papers from seven randomized trials compared exercise protocols that allow or encourage pain against protocols that stay below it, in 385 people with chronic musculoskeletal pain. The painful protocols won on short-term pain by a standardized mean difference of 0.27 (Smith 2017).
The margin was small and the interval ran from 0.05 to 0.54. At medium and long-term follow-up the two approaches were indistinguishable. What the review settled was the safety question.
A tendon trial made the same point under harder loading. Thirty-eight patients with Achilles tendinopathy were randomized either to keep running and jumping under a pain-monitoring model or to stop for six weeks (Silbernagel 2007). Scores on the Swedish Victorian Institute of Sports Assessment questionnaire rose from 57 to 85 in the group that kept loading, and from 57 to 91 in the group that rested. No negative effect of continued loading could be demonstrated.
Two thresholds, and the gap between them
Two separate thresholds sit inside every painful movement. The first is tissue tolerance, the point at which tissue would actually be strained or injured. The second is the pain threshold, the point at which the nervous system activates protection. In a healthy system the pain threshold sits well below tissue tolerance, giving a wide margin of safety.
Our model reads that gap as the design of the system rather than an accident of it. Because pain is built to stop injury before it happens, discomfort usually arrives long before anything approaches real tissue failure. The two trials above are what the gap looks like when someone measures it. People loaded into pain for weeks and their tissue held.
02Findings
What the research shows
From two meta-analyses, a systematic review of fear-avoidance beliefs, and four randomized trials of loading, activity and exposure.
03Fear turns the volume up
Pain-related fear predicts disability better than the injury does
Pain output is flexible, and the nervous system turns it up or down according to context. Expectation is one of those contexts, and it is the one a clinician can change. A nervous system protecting healing tissue raises sensitivity deliberately, so that gentle loading feels tender. The tissue state has not changed. The volume has.
Pain is therefore better understood as produced by the nervous system than transmitted from the tissues like a fixed signal on a wire. Pain Is Not Tissue Damage sets out the modulation machinery and the experiments that isolated it.
Lethem and colleagues named the behavioral half of that machinery in 1983, calling it the fear-avoidance model. Vlaeyen and Linton reviewed two decades of work on it (Vlaeyen 2000). Fear of pain produces two extreme responses. Confrontation reduces fear over repetitions. Avoidance maintains or worsens it, and can generate something close to a phobic state.
Fear-avoidance beliefs also predict outcome. In low back pain of under six months, high beliefs tracked with more pain, more disability and less return to work across 17 randomized trials (Wertli 2014). In pain that had already turned chronic the picture was mixed. Two studies of 339 patients found the association and three of 832 found none.
Where the cost of guarding actually shows up
The disuse hypothesis, which held that guarding drives chronicity through lost fitness, was not confirmed. Verbunt and colleagues reviewed daily activity levels and physical fitness in chronic low back pain against healthy controls (Verbunt 2003). Activity levels came back either lower or comparable. Fitness came back either lower or comparable.
The fit patients were more often still employed, and employment itself involves physical activity. Where tissue has genuinely gone unused, reconditioning restores its function. The reliable cost of avoidance is the shrinking list of movements a person will attempt.
04Overprotection after healing
Protection outlasts repair, and a flare is judged by what it leaves behind
The nervous system does not update its settings on the same schedule as the tissue. Long after repair, the system can go on behaving as though danger is present. Previous experiences, alarming interpretations of imaging, and understandable fear all argue that the body remains fragile. The system answers by lowering the threshold at which pain occurs.
The result is overprotection rather than ongoing tissue damage. The body is sound, but the alarm is set too sensitively. That distinction matters, because it reframes persistent discomfort as a recalibration challenge rather than evidence of a broken structure. Central Sensitization sets out the mechanism underneath it.
What separates a flare from a setback
A flare is a rise in symptoms that settles. The distinction a clinician draws is between a reaction that leaves function where it was and one that lowers it. Across seven trials and 385 participants, loading into pain cost nothing on function or disability at any follow-up point (Smith 2017).
The tendon data reads the same way. Six weeks of continued running and jumping on a painful Achilles left twelve-month questionnaire scores at 85, against 91 for the group that rested (Silbernagel 2007). The reaction was loud and the trajectory held. What a flare leaves behind is the measurement, and the noise it made on the day is not.
05Measuring fear of movement
Kinesiophobia has a questionnaire, a score, and a change threshold
Kinesiophobia is fear of movement and of reinjury, and it is measured rather than guessed at. The Tampa Scale for Kinesiophobia is among the most frequently used measures of pain-related fear in back pain (Woby 2005).
Woby and colleagues tested the English version in chronic low back pain patients. Four items showed low item-total correlations and four had skewed response patterns. Removing the six poorest left an 11-item scale, the TSK-11. It matched the original 17-item version on internal consistency, at alpha 0.79 against 0.76, and on test-retest reliability, at 0.81 against 0.82.
The number a clinician can use is the change threshold. A reduction of at least four points on either version maximized the chance of correctly identifying a real drop in fear of movement. That gives a scored variable to track between visits rather than an impression.
What the score is made of
The scale splits into two factors, somatic focus and activity avoidance. Roelofs and colleagues tested that structure in four conditions: work-related upper extremity disorders, chronic low back pain, fibromyalgia and osteoarthritis (Roelofs 2007). It held in all four, and across Dutch, Swedish and Canadian samples. Both factors were modestly but significantly related to disability.
Two people reporting the same pain intensity can hold very different expectations about what movement will do to them. The score names that difference before treatment starts.
06Graded activity
Graded activity raises the workload on a schedule and ignores the daily symptom
Graded activity progresses on a quota. The person completes the day’s allotment whether the day feels good or bad, which removes pain as the signal governing effort.
The method was tested in 1992 on industrial blue-collar workers who had been sick-listed for eight weeks with subacute nonspecific mechanical low back pain. Of 103 patients, 51 went to the activity group and 52 to control (Lindstrom 1992). The program ran in four parts: measurement of functional capacity, a workplace visit, back school education, and an individual submaximal exercise program that rose gradually.
The activity group returned to work significantly earlier. In the second follow-up year they took 12.1 weeks of sick leave against 19.6 weeks in the control group. One activity patient received a permanent disability pension, against four controls. The median number of physical therapist appointments before return to work was five.
Pooled across later trials the effect is real and modest. Macedo and colleagues reviewed 15 randomized trials in 1,654 patients with persistent low back pain (Macedo 2010). Against a minimal intervention, graded activity gave 6.2 points of 100 on short-term pain and 6.5 on short-term disability, with intermediate-term gains of 5.5 and 3.9.
It gave nothing over other forms of exercise. Six trials made that comparison and none of the pooled effects reached significance. Progressive loading is the ingredient, and the brand name on the progression is not.
Staying active beat resting, and it beat prescribed exercise too
The plainest version of the same idea was tested in Helsinki. 186 employees with acute nonspecific low back pain were randomized to two days of bed rest, to back-mobilizing exercises, or to ordinary activities as tolerated (Malmivaara 1995). The ordinary-activity group recovered better at three and twelve weeks.
They did better on duration of pain, pain intensity, lumbar flexion and days absent from work. Recovery was slowest after bed rest, and total costs did not differ.
07Graded exposure
Graded exposure works on the expectation of harm rather than on the workload
Graded exposure starts from what a person believes an activity will do to them. Ranked from least feared to most, activities are reintroduced gently in that order until the outcome corrects the expectation.
The head-to-head trial ran in 85 patients with disabling nonspecific chronic low back pain who reported at least moderate pain-related fear. They were randomized to exposure in vivo or to operant graded activity (Leeuw 2008). Exposure was clearly superior at reducing pain catastrophizing and the perceived harmfulness of activities.
On the outcomes that mattered most it drew level. Functional disability and main complaints improved about equally, with the difference favoring exposure and falling just short of significance. Pain intensity and daily activity levels did not differ. Exposure was no better in the most fearful subgroup either.
The mediation analysis explains the shape of that result. What effect exposure did have on disability and main complaints ran through the drop in catastrophizing and perceived harmfulness. Belief was the route it took, and belief was not the only route available.
The pooled evidence puts the two methods level. Four trials compared them directly and no pooled effect reached significance. Two trials compared graded exposure with a waiting list and found nothing there either (Macedo 2010). Both methods progress. One raises the quota and the other raises the expectation, and a clinician chooses by what the person is holding.
08Reframe, remap, relearn
Confident recovery runs through three stages
The first stage is to reframe, replacing an inaccurate understanding of pain with a better one. Painful movement does not always mean injury, the body is resilient, and moving within a sensible range is safe. That shift reduces fear and encourages activity, which is why education counts as treatment rather than a preamble to it.
The second stage is to remap. Varied movement, changes of position and gentle challenge restore the sensory awareness the nervous system uses to build its picture of the body. The third stage is to relearn, rebuilding smooth and confident movement patterns for daily life. Neuroplasticity and Rehab carries the cortical detail.
Dosing the return
When pain thresholds are very low, formal exercise can sit entirely inside the protective zone and accomplish little. Early efforts focus instead on restoring movement, improving alignment and reintroducing normal body use. As protection eases, activity is nudged slightly above the current threshold to demonstrate safety, expand tolerance and build confidence, and structured exercise then becomes far more effective.
That is graded activity described in plain clinical terms, and 12.1 weeks of sick leave against 19.6 is what it bought in the trial that first tested it (Lindstrom 1992). A person who expects a movement to injure them will not produce that movement, however well the quota is written. A clear and personal explanation carries more weight than a generic one.
When people learn that hurt does not necessarily equal harm, fear begins to loosen its grip and movement returns. None of this replaces clinical judgment about when different care is warranted.
09Claims removed from this page
Three claims from the earlier version were removed
The statement that most connective tissue heals within several weeks carried no source and named no tissue, and healing times differ widely between tissues. The measured recovery curve for low back pain replaces that claim (Menezes Costa 2012), and Conservative First carries it. The claim that a long stretch of guarding produces deconditioning was cut back to what the evidence supports. A review of activity levels and fitness in chronic low back pain did not confirm disuse (Verbunt 2003).
The list of what the nervous system is looking for, given as safety, expertise, individual attention and meaning, came off because nothing cited here measured it. A gold pull-quote block came off as well, and its sentence now runs inside the prose where it can be read as the claim it is.
10Judging the response
What the Unified Model of Tone claims about hurt, harm and flares
Everything above is established science, including the comparisons that came back null. What follows is our model’s reading, stated as ours rather than drawn from the papers cited.
An input interacting with a tone creates an outcome. There is no such thing as an input acting upon an empty body. The same graded program delivered to two people with the same diagnosis meets two different states, and the outcomes differ for that reason. Six trials found graded activity indistinguishable from other exercise (Macedo 2010), which is what a fixed input averaged across an unstratified sample produces.
Why worse does not mean working
Our model refuses the move that reads a bad reaction as proof of progress. Worsening after an intervention can equally mean the input was inappropriate, excessive, mistimed or harmful. A flare is information about the match between the input and the state. On its own it carries no verdict at all.
Response is judged through function and durability, through physiological stability and adverse effects, and through whether the person’s capacity is expanding or narrowing. The drama of an immediate reaction is not on that list. The Achilles trial is the closest published version of that discipline, and no negative effect of the continued loading could be demonstrated (Silbernagel 2007).
The prediction
The most informative variable is frequently not the starting number but the ability to change appropriately and return. From that follows a claim the fear-avoidance literature does not make. Our model holds that fear of movement, available range, sensory gain and recovery speed are readings of one organization rather than four independent traits.
Four measures make that testable in the same people. They are the TSK-11 score, active lumbar flexion in degrees, pressure pain threshold away from the painful region, and time to return to baseline pain after a standardized load test. Our model predicts they share one underlying factor across a graded activity program.
Our model further predicts the direction of change under an input that restores regulation. People who begin hypersensitive on pressure pain threshold and people who begin hyposensitive both move toward the middle, and the spread across the group narrows.
This is a claim about how protection is organized rather than a claim about what treatment does. If the TSK-11 score, active lumbar flexion, pressure pain threshold away from the painful region and recovery time after a load test are shown to move together, the unification claim is confirmed.
11The tone reading
How hurt and harm express tone
Every topic in this library expresses all of tone. In hurt and harm three aspects carry the signature, because the same graded program moves two people differently depending on what each expects it to do to them.
Prediction
Expected harmfulness governs what a person attempts. Exposure in vivo cut perceived harmfulness more than graded activity did, and that drop carried the gain in function.
Constraint
Avoidance narrows the movements a person will attempt, and the narrowing holds even where fitness testing finds no measurable deconditioning across the group.
Load
Progression runs on a quota rather than on symptoms. Graded activity cut sick leave in the second follow-up year from 19.6 weeks to 12.1.
The remaining foundations run through hurt and harm as well. Time course: fear falls across repeated contact with a feared movement and climbs again when the movement is dropped. Gain: the pain threshold sits below tissue tolerance, and how far below decides what a person can load. Set point: the resting level of protection decides how much movement it takes before the alarm sounds. Input quality: poor proprioceptive information leaves the system less to work with when it estimates whether a movement is dangerous. Coupling: held breath, muscular guarding and hesitation arrive together at the moment a feared movement begins. Oscillation: symptoms that swing on a daily rhythm are reporting the regulator rather than the tissue. These are readings of one organization rather than 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
Every page in this section that asks a person to move rests on the distinction between hurt and harm.
Why the pain threshold drops in the first place, measured with quantitative sensory testing.
Where fear of movement sits among the psychological factors that predict a poor recovery.
Why no single muscle explains recovery, and why varied movement outperforms bracing.
The tissue where loading into symptoms is standard care, with the heavy slow resistance protocols.
The remapping stage described at the level of cortical representation and repetition.
The input the nervous system uses to keep its map of the body current, and what stillness costs it.
The claim underneath this one. Pain is an output the nervous system produces, and the emergency clinic data, the imaging data and the education trials are what establish it.
13Frequently asked
Questions patients ask about hurt, harm and moving again
Does hurting during movement mean I am causing damage?
Usually not. Pain arrives at a threshold that sits below the point where tissue would actually be strained, so discomfort normally shows up long before anything is at risk. Seven randomized trials in 385 people with chronic musculoskeletal pain compared exercise that allowed pain against exercise that stayed below it. The painful protocols were slightly better in the short term and no worse later. Pain during therapeutic exercise did not block a good outcome in any of them.
Should exercise be painful?
It does not have to be, and it is allowed to be. The pooled advantage of protocols that permit pain is a standardized mean difference of 0.27 on short-term pain, which is small, and it disappears at medium and long follow-up. What the comparison establishes is safety rather than superiority. In one Achilles tendinopathy trial, patients who kept running and jumping under a pain-monitoring model improved from 57 to 85 on a twelve-month questionnaire with no negative effect from the loading.
What is graded activity?
Graded activity is a program that progresses on a quota instead of on symptoms. A starting workload is set from measured functional capacity, low enough to complete on a poor day, and increases follow a written schedule. The person completes the quota whether the day feels good or bad. In the 1992 trial that established it, 103 sick-listed workers were randomized, and the activity group took 12.1 weeks of sick leave in the second follow-up year against 19.6 weeks.
How is graded exposure different?
Graded exposure targets the expectation rather than the workload. Activities are ranked by how harmful the person expects them to be, from least feared to most, then reintroduced in that order until the outcome corrects the belief. In a trial of 85 chronic low back pain patients with at least moderate fear, exposure cut catastrophizing and perceived harmfulness more than graded activity did, while disability improved about equally in both groups. Roughly half of each group reported a clinically relevant improvement, and four pooled trials put the two methods level.
What is kinesiophobia and how is it measured?
Kinesiophobia is fear of movement and of reinjury, and it is measured rather than guessed at. The Tampa Scale for Kinesiophobia is the usual instrument, and its short form keeps 11 of the original 17 items while matching the long version on reliability. The scale splits into two factors, somatic focus and activity avoidance, and that structure held across four pain conditions and three countries. A drop of at least four points is the change that best identifies a real reduction in fear.
If treatment makes me feel worse at first, does that mean it is working?
Not by itself. Worsening after an input can equally mean the input was inappropriate, excessive, mistimed or harmful, so the reaction alone settles nothing. A clinician judges the response through function and durability, through physiological stability and adverse effects, and through whether capacity is expanding or narrowing. Loading into pain across seven trials left function and disability no worse at any point. Six weeks of running on a painful Achilles left twelve-month scores at 85, against 91 for rest. Those measurements settle the question rather than the size of the reaction.
What does the Unified Model of Tone say about hurt and harm?
That an input interacting with a tone creates the outcome, which is why the same graded program lands differently on two people with the same diagnosis. Six trials found graded activity indistinguishable from other exercise, which is what a fixed input averaged across an unstratified sample produces. From that the model predicts that fear of movement, available range, pressure pain threshold and recovery time share one underlying factor, with compensation deciding how far each one moves. That prediction concerns how protection 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