Why Patients Rapidly Change After Initiating Treatment: What We Think We Know and What We Likely Don’t

By: Chad Cook, PT, PhD, FAPTA

Background: The completion of our SS‑MECH trial [1] in the fall of 2025 marked the 20th clinical trial for which I had direct access to the data. Over the years, I have come to view data as a narrative, one that cannot be fully captured by summary statistics, tables, or figures alone. Across every trial I have been involved in, a consistent pattern emerges: patients often experience rapid improvement soon after initiating treatment, including those with persistent, complex, and longstanding symptoms. This pattern has appeared in every single trial.

Early Change is Prognostic: Several studies have shown that early symptom improvement, particularly when consistent across consecutive sessions, is a strong prognostic indicator of favorable clinical outcomes [2–4]. These meaningful shifts typically emerge within the first one to three treatment sessions, or over the initial one to two weeks of care. Importantly, early improvements must be substantial, often 30% or more from baseline, to carry prognostic significance, as smaller fluctuations are more likely attributable to measurement variability. The pattern is remarkably consistent across all treatment types and appears to reflect genuine patient‑level change rather than the specific effects of any single intervention. This includes studies involving psychologically-informed care, which is not traditionally associated with rapid symptom change, suggesting that early improvement reflects a broader, trans‑theoretical signal of treatment responsiveness [5,6].

Why Outcomes Change Quickly in Musculoskeletal (MSK) Pain: Several complementary theories help explain why rapid clinical improvement can occur even in individuals who have experienced symptoms for years. Contemporary pain science emphasizes the role of predictive processing, in which pain is understood as the brain’s best inference about potential bodily threat. The moment treatment begins, several factors can immediately recalibrate these predictions, such as: 1) reduced uncertainty (“Someone finally understands what’s happening to me”); 2) the introduction of new safety cues (reassurance, clear explanations, a supportive clinical environment); and 3) the formation of updated expectancies (“Maybe this can improve”). Together, these elements shift the brain’s threat appraisal, which can quickly down‑regulate pain intensity. As the threat diminishes, functional assumptions and behavioral patterns begin to reorganize, creating a cascade of early, meaningful change.

Secondly, descending inhibitory pathways can activate quickly [7,8]. The central nervous system has built‑in systems that can turn down nociceptive signaling in the periaqueductal gray, the rostral ventromedial medulla, and prefrontal and anterior cingulate cortex involvement in top‑down modulation. Therapeutic interactions, reassurance, movement exposure, and even simple attention shifts can activate these pathways within minutes. Further, when pain modulation occurs during a session, it reflects rapid engagement of the nervous system’s endogenous pain‑inhibitory mechanisms [7]. When a patient experiences immediate symptom relief, whether through movement, education, manual therapy, or psychologically informed strategies, it signals that descending inhibitory pathways are intact and responsive. Again, this is likely a patient‑level function, not an interventional reflection.

Thirdly, expectancy and contextual effects are potentially biologically influential contributors. Contextual effects (including contextual benefits and harms) are not “fake”; they involve measurable neurochemical changes involving endogenous opioids, endocannabinoids, dopamine, and reduced amygdala activation [9,10]. These systems can change pain perception extremely quickly. Chronic pain does not block these mechanisms; in fact, many chronic pain patients respond strongly because they’ve been in a prolonged state of uncertainty or fear.

Fourthly, rapid changes in pain perception are accompanied by rapid changes in movement [11,12]. In our SS‑MECH trial data [1], we observed a 13% improvement in a composite measure of overall range of motion in both treatment groups from baseline after only 1 visit and 1 week of home exercises. We likely saw these changes because we taught safe movement strategies, provided reassurance, allowing the nervous system to relax, protective bracing, altered motor patterns (maybe), and fear‑based movement avoidance. These changes are neuromuscular, not structural, so they can happen very fast.

This is what I heard in a pain neuroscience course-it must be correct: I recognize that this may sound like the familiar narrative presented in many pain neuroscience courses. It’s coherent, well‑constructed, and intuitively appealing. But is it correct? To explore that question, it’s worth examining four key counterpoints to these commonly cited explanations.

Point One: Predictive processing models of pain are influential, but they remain theoretical frameworks, not empirically validated causal mechanisms. Critics will argue that predictive coding is difficult to falsify because it can explain almost any outcome [13]. There is limited direct evidence that clinical encounters rapidly alter cortical predictions in a way that explains early symptom change. Many rapid improvements could be explained by regression to the mean, natural variability, or measurement noise, rather than shifts in threat prediction. In other words, predictive processing may be one explanation, but it is not the only one, and it is not yet empirically established as the explanation [14].

Point Two: Rapid activation of descending inhibition is plausible but not universally demonstrated [7]; responsiveness varies widely across individuals and conditions. Many patients with chronic pain have impaired descending inhibition, yet still show early improvement in clinical trials. In-session pain reduction may reflect peripheral changes, movement variability, or simple symptom fluctuation, not necessarily descending inhibition. Thus, while descending inhibition is a real system, its role in early clinical change is not definitively established.

Point Three: Contextual and expectancy effects are real, but their magnitude is often overstated [15]. A recent meta-analysis quantified their contributions to meaningful thresholds of clinical outcomes and estimated approximately 30% [16]. Many placebo studies show improvements of only 5–10 points on a 100‑point scale. Not all patients respond strongly to contextual cues, especially those with entrenched chronic pain. Expectancy effects are highly variable and depend on personality, prior experiences, and cultural context. Thus, contextual effects contribute, but they may not fully explain rapid, clinically meaningful change.

Point Four: Early change may reflect statistical and methodological artifacts [17]. A strong critique is that early change may not reflect any biological mechanism. Instead, it may reflect: 1) regression to the mean (patients seek care when symptoms peak); 2) natural symptom variability (MSK pain fluctuates day to day); 3) Hawthorne effects (i.e., patients improve simply because they are being observed); 4) measurement error (no measurement device is perfect); and/or 5) response bias (patients want to please the clinician). These explanations require no neurobiological mechanism and are well‑documented in clinical research.

What does this mean as a clinician? The perspectives in this blog illustrate why rapid improvement in MSK pain conditions is plausible, commonly observed, yet not fully understood. They are likely a reflection of multiple neurobiological, cognitive, and contextual systems that can shift quickly. Each proposed mechanism associated with predictive processing has important limitations, and none fully accounts for the phenomenon. This underscores the need for a more cautious, evidence‑based interpretation of early clinical change, one that recognizes both the genuine biological capacity for rapid improvement and the methodological artifacts that can mimic it.

Layperson summary: Pain can drop quickly when the brain feels safer, more certain, and more hopeful. The nervous system has built‑in “pain‑dimming” systems that can switch on within minutes. Supportive interactions, clear explanations, and safe movement can trigger real chemical changes that reduce pain fast.  When pain eases, movement improves quickly as compensatory patterns reduce. Some early improvement is real biology; some is just normal symptom fluctuation.

References:

1. Cook CE, O’Halloran B, McDevitt A, Keefe FJ. Specific and shared mechanisms associated with treatment for chronic neck pain: study protocol for the SS-MECH trial. J Man Manip Ther. 2024 Feb;32(1):85-95.
2. Donaldson M, Petersen S, Cook C, Learman K. A Prescriptively Selected Nonthrust Manipulation Versus a Therapist-Selected Nonthrust Manipulation for Treatment of Individuals With Low Back Pain: A Randomized Clinical Trial. J Orthop Sports Phys Ther. 2016 Apr;46(4):243-50.
3. Cook C, Petersen S, Donaldson M, Wilhelm M, Learman K. Does early change predict long-term (6 months) improvements in subjects who receive manual therapy for low back pain? Physiother Theory Pract. 2017 Sep;33(9):716-724.
4. Cook C, Lawrence J, Michalak K, Dhiraprasiddhi S, Donaldson M, Petersen S, Learman K. Is there preliminary value to a within- and/or between-session change for determining short-term outcomes of manual therapy on mechanical neck pain? J Man Manip Ther. 2014 Nov;22(4):173-80.
5. Schibbye P, Ghaderi A, Ljótsson B, Hedman E, Lindefors N, Rück C, Kaldo V. Using early change to predict outcome in cognitive behaviour therapy: exploring timeframe, calculation method, and differences of disorder-specific versus general measures. PLoS One. 2014 Jun 24;9(6):e100614.
6. Beard JIL, Delgadillo J. Early response to psychological therapy as a predictor of depression and anxiety treatment outcomes: A systematic review and meta-analysis. Depress Anxiety. 2019 Sep;36(9):866-878.
7. Yarnitsky D. Conditioned pain modulation (CPM): what do we know? Pain. 2015;156 Suppl 1:S93 S97.
8. Ossipov MH, Morimura K, Porreca F. Descending pain modulation and chronification of pain. Curr Opin Support Palliat Care. 2014;8(2):143 151.
9. Colloca L, Barsky AJ. Placebo and nocebo effects. N Engl J Med. 2020;382(6):554 561.
10. Benedetti F. Placebo effects: from the neurobiological paradigm to translational implications. Neuron. 2014;84(3):623 637.
11. Hodges PW, Smeets RJ. Interaction between pain, movement, and physical activity: short term benefits, long term consequences, and targets for treatment. Clin J Pain. 2015;31(2):97 107.
12. van Dieën JH, Flor H, Hodges PW. Low back pain patients learn to adapt movement patterns. Nat Rev Rheumatol. 2017;13(12):731 740.
13. Sullivan MD, Ballantyne JC. Must we reduce pain intensity to treat chronic pain? Pain. 2016;157(1):65 69.
14. Louw A, Goldrick S, Bernstetter A, Van Gelder LH, Parr A, Zimney K, Cox T. Evaluation is treatment for low back pain. J Man Manip Ther. 2021 Feb;29(1):4-13.
15. Hróbjartsson A, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database Syst Rev. 2010;(1):CD003974.
16. Saueressig T, Owen PJ, Pedder H, Tagliaferri S, Kaczorowski S, Altrichter A, Richard A, Miller CT, Donath L, Belavy DL. The importance of context (placebo effects) in conservative interventions for musculoskeletal pain: A systematic review and meta-analysis of randomized controlled trials. Eur J Pain. 2024 May;28(5):675-704.
17. Barnett AG, van der Pols JC, Dobson AJ. Regression to the mean: what it is and how to deal with it. Int J Epidemiol. 2005;34(1):215 220.