Short-Form McGill Pain Questionnaire (SF-MPQ-2)

Abstract

The Short-Form McGill Pain Questionnaire (SF-MPQ-2) is a multidimensional self-report measurement tool engineered to assess the sensory, affective, and neuropathic dimensions of chronic and acute pain. Developed by Robert H. Dworkin, Ronald Melzack, and colleagues under the auspices of the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) in 2009, the SF-MPQ-2 is an expanded and revised version of Melzack’s seminal 1987 Short-Form McGill Pain Questionnaire (SF-MPQ). The primary objective of the revision was to incorporate relevant symptoms characteristic of neuropathic pain conditions and to replace the original 4-point categorical intensity scale with a 0 to 10 numerical rating scale (NRS), thereby enhancing psychometric sensitivity, measurement precision, and responsiveness to therapeutic change.

The instrument comprises 22 standardized pain descriptors distributed across four empirically validated subscales: Continuous Pain (6 items), Intermittent Pain (6 items), Predominantly Neuropathic Pain (6 items), and Affective Descriptors (4 items). Each item is rated on an 11-point numerical rating scale ranging from 0 (“none”) to 10 (“worst possible”). Extensive psychometric evaluations across diverse chronic pain cohorts—including peripheral neuropathy, low back pain, osteoarthritis, cancer-related pain, and fibromyalgia—demonstrate exceptional internal consistency reliability. Published Cronbach’s alpha coefficients consistently range from 0.85 to 0.92 across the subscales and exceed 0.90 for the total score. Confirmatory factor analysis (CFA) reliably supports the four-factor oblique structural model, exhibiting superior fit relative to unidimensional or alternative hierarchical solutions. The SF-MPQ-2 serves as an indispensable instrument in clinical pharmacology trials, epidemiological investigations, and specialized pain management protocols worldwide.

Keywords

Short-Form McGill Pain Questionnaire, SF-MPQ-2, pain measurement, neuropathic pain, sensory pain descriptors, affective pain, psychometrics, Gate Control Theory, IMMPACT, Ronald Melzack

Authors

The revised and expanded SF-MPQ-2 was developed through an international collaborative effort led by the IMMPACT consensus group, building directly upon the original conceptual framework formulated by Professor Ronald Melzack.

• Ronald Melzack, Ph.D.: Emeritus Professor of Psychology, Department of Psychology, McGill University, Montreal, Quebec, Canada. (Pioneer of the Gate Control Theory and creator of the original MPQ and SF-MPQ). Inquiries: [email protected].
• Robert H. Dworkin, Ph.D.: Professor of Anesthesiology, Neurology, and Oncology, School of Medicine and Dentistry, University of Rochester Medical Center, Rochester, NY, USA. (Co-founder and Director of IMMPACT).
• Dennis C. Turk, Ph.D.: Professor of Anesthesiology and Pain Medicine, University of Washington, Seattle, WA, USA.
• Dennis A. Revicki, Ph.D.: Senior Vice President, Outcomes Research, Evidera (formerly Center for Health Outcomes Research, United BioSource Corporation), Bethesda, MD, USA.
• Collaborating IMMPACT Investigators: G. Harding, K. S. Coyne, S. Peirce-Sandner, D. Bhagwat, D. Everton, L. B. Burke, P. Cowan, J. T. Farrar, S. Hertz, M. B. Max, and B. A. Rappaport.

Purpose

Pain is an intrinsically complex, multidimensional, subjective experience encompassing sensory discriminative qualities, emotional affective states, and cognitive evaluations. For decades, standard assessment protocols in both clinical practice and clinical trial research relied heavily on unidimensional pain intensity measures, such as the Visual Analogue Scale (VAS) or the 11-point Numeric Rating Scale (NRS) of pain severity. Although unidimensional scales capture the magnitude of distress or sensory intensity, they fail entirely to characterize the qualitative phenotypes of pain, discriminate between underlying neurobiological mechanisms, or capture changes in specific symptom clusters following pharmacological or behavioral interventions.

The original Short-Form McGill Pain Questionnaire (Melzack, 1987) provided a brief 15-item inventory derived from the 78-word long-form MPQ. However, clinical and pharmacological trials increasingly revealed critical limitations in the 1987 instrument: it lacked adequate representation of symptoms characteristic of neuropathic pain (such as numbness, tingling, allodynia, and electric-shock sensations), and its 4-point Likert-style verbal descriptor scale (0 = none, 1 = mild, 2 = moderate, 3 = severe) lacked the continuous statistical granularity and measurement sensitivity needed to detect subtle therapeutic responses in modern clinical trials.

The explicit purpose of the SF-MPQ-2 is to provide an efficient, psychometrically robust, self-administered questionnaire that:

• Accurately assesses and distinguishes between both neuropathic and non-neuropathic chronic pain symptoms within a single unified framework.
• Quantifies four distinct phenotypic pain dimensions: continuous steady pain, intermittent paroxysmal pain, neuropathic sensations, and affective-motivational distress.
• Measures clinical trial outcomes in accordance with IMMPACT recommendations, utilizing an expanded 0 to 10 response format that aligns with international regulatory standards (e.g., FDA and EMA guidance for patient-reported outcomes).
• Facilitates differential diagnosis, informs mechanism-based treatment stratification, and monitors patient-specific trajectories in diverse settings including rheumatology, neurology, oncology, orthopedic surgery, and interventional pain rehabilitation.

Psychological Construct

The psychological and physiological construct evaluated by the SF-MPQ-2 is multidimensional pain experience. Rather than conceptualizing pain as a mere passive sensory read-out proportional to peripheral tissue injury, the construct operationalizes pain as a dynamic emergent property of distributed neural processing. The SF-MPQ-2 measures four intercorrelated yet distinct latent constructs:

1. Continuous Pain

This subscale assesses steady, tonic, background sensory discomfort that persists without substantial fluctuation. It captures deep somatic and visceral sensations reflecting sustained nociceptive signaling, tissue inflammation, or persistent central sensitization. Descriptors representing this dimension include throbbing, cramping, gnawing, aching, heavy, and tender. Patients experiencing conditions such as rheumatoid arthritis, muscle spasm syndromes, and visceral inflammatory disorders typically endorse elevated scores on this continuous sensory dimension.

2. Intermittent Pain

The intermittent pain dimension captures paroxysmal, episodic, sharp sensations that punctuate baseline pain or occur as sudden lancinating flares. This dimension encompasses sharp mechanical insults, brief neural paroxysms, or visceral colic. The specific items defining this construct are shooting, stabbing, sharp, splitting, electric-shock pain, and piercing. High scores on this subscale reflect intermittent mechanical instability, neurovascular compression (as observed in trigeminal neuralgia), or acute radicular impingement.

3. Predominantly Neuropathic Pain

Introduced systematically in the SF-MPQ-2, this dimension operationalizes abnormal sensory phenomena directly resulting from a lesion or disease affecting the somatosensory nervous system. The construct captures positive sensory phenomena (paresthesia, dysesthesia, hyperalgesia, allodynia) and negative sensory phenomena (hypoesthesia). The items comprise hot-burning, cold-freezing, pain caused by light touch (dynamic mechanical allodynia), itching/latching, tingling or ‘pins and needles’, and numbness. This scale demonstrates discriminant utility in identifying peripheral neuropathies (e.g., diabetic distal symmetric polyneuropathy, postherpetic neuralgia, chemotherapy-induced peripheral neuropathy) and central post-stroke pain syndromes.

4. Affective Descriptors

This dimension operationalizes the emotional, motivational, and unpleasant distress component intrinsic to the pain experience. It reflects the suffering, unpleasantness, perceived threat, and psychological fatigue generated by unremitting nociceptive barrage. The construct is evaluated through four items: tiring-exhausting, sickening, fearful, and punishing-cruel. This dimension demonstrates robust correlations with psychological constructs including pain catastrophizing, clinical depression, somatic anxiety, and state of helplessness.

Theoretical Framework

The SF-MPQ-2 is rooted in the revolutionary theoretical models of pain perception articulated by Ronald Melzack and Patrick D. Wall, specifically the Gate Control Theory of Pain (Melzack & Wall, 1965) and the subsequent Neuromatrix Model of Pain (Melzack, 1990, 1999, 2005).

The Gate Control Paradigm

Prior to the Gate Control Theory, pain was viewed through a simplistic Cartesian specificity model: dedicated peripheral pain fibers projected linearly to a specialized “pain center” in the brain, with perceived pain directly proportional to tissue damage. Melzack and Wall proposed instead that neural mechanisms in the dorsal horn of the spinal cord (specifically the substantia gelatinosa) act like a dynamic gate that modulates the transmission of nerve impulses from peripheral fibers (large-diameter A-beta fibers versus small-diameter A-delta and C fibers) to central transmission cells. Crucially, this gating mechanism is regulated not only by descending inhibitory or facilitatory influences from the brainstem, but also by cognitive, emotional, and attentional processes originating in the cerebral cortex.

The Neuromatrix Model

Expanding on the Gate Control Theory, Melzack formulated the Neuromatrix Model to account for phantom limb pain, chronic regional pain syndromes, and the complex dissociation between somatic pathology and subjective pain intensity. Melzack postulated that pain is produced by the output of a widely distributed neural network—the body-self neuromatrix—which integrates inputs from:

• Sensory-discriminative systems: projecting through the spinothalamic tracts to the primary and secondary somatosensory cortices, decoding stimulus location, duration, quality, and physical intensity.
• Affective-motivational systems: processing information via the reticular formation, limbic structures, amygdala, and anterior cingulate cortex (ACC), governing autonomic responses, emotional valence, aversion, and escape behaviors.
• Cognitive-evaluative systems: involving prefrontal cortical networks, memory stores, cultural meanings, expectations, and attentional focus.

The cyclical processing and synthesis of these inputs generate a characteristic neurochemical profile called the neurosignature, which yields the conscious experience of pain. The SF-MPQ-2 directly operationalizes this tripartite neurobiological architecture by decomposing subjective pain into sensory (continuous, intermittent), neuropathic, and affective subscales, providing a structured psychometric interface with the underlying neuromatrix components.

Validity

The psychometric validity of the SF-MPQ-2 has been corroborated through rigorous multi-center clinical trials and cross-cultural validation investigations.

Construct and Convergent Validity

In the landmark validation study conducted by Dworkin et al. (2009) involving 882 patients presenting with chronic neuropathic (diabetic peripheral neuropathy, postherpetic neuralgia, radiculopathy) and non-neuropathic (osteoarthritis, chronic low back pain) conditions, convergent validity was established through high correlations with established pain metrics. The SF-MPQ-2 total score correlated strongly with the Brief Pain Inventory (BPI) average pain item (r = 0.69) and the Pain DETECT screening tool (r = 0.73). Subscales exhibited logical convergent patterns: the Predominantly Neuropathic Pain subscale correlated significantly higher with the Pain DETECT score (r = 0.72) than did the Continuous Pain subscale (r = 0.52), verifying that the new items accurately converge on neuropathic phenomenology.

Discriminant Validity

Discriminant validity was demonstrated by the questionnaire’s ability to differentiate clinical diagnostic categories known to possess divergent pathophysiological etiologies. Patients with physician-confirmed peripheral neuropathies scored significantly higher on the Predominantly Neuropathic subscale (mean = 3.65, SD = 2.14) compared to patients with mechanical osteoarthritis (mean = 1.94, SD = 1.62; p < 0.001). Conversely, osteoarthritis patients exhibited predominant elevations on the Continuous Pain and Intermittent Pain subscales. Furthermore, the Affective subscale demonstrated divergent validity from sensory measures by showing unique, robust associations with the Beck Depression Inventory (BDI-II) and the Pain Catastrophizing Scale (PCS), demonstrating that the affective items track emotional distress independently of raw sensory nociceptive intensity.

Predictive and Responsiveness Validity

The SF-MPQ-2 displays marked sensitivity to change (responsiveness) in pharmacological clinical trials. In randomized controlled trials assessing pregabalin, duloxetine, and topical analgesics, reductions in SF-MPQ-2 total and subscale scores correlated robustly with the Patient Global Impression of Change (PGIC) (Maruo et al., 2014). Standardized response means (SRMs) for the SF-MPQ-2 in effective analgesic treatment arms consistently range between 0.70 and 1.10, indicating moderate-to-large effect sizes for tracking clinical improvement over time.

Reliability

The SF-MPQ-2 demonstrates exceptional internal consistency and temporal test-retest reliability across clinical populations and language adaptations.

Internal Consistency Reliability

In the initial developmental cohort evaluated by Dworkin et al. (2009), Cronbach’s alpha coefficients across all clinical samples demonstrated high internal consistency:

• Continuous Pain subscale: α = 0.893
• Intermittent Pain subscale: α = 0.875
• Predominantly Neuropathic Pain subscale: α = 0.917
• Affective Descriptors subscale: α = 0.857
• Total Score (22 items): α = 0.907

Subsequent international validation studies have replicated these findings. For instance, the Japanese validation study conducted by Maruo et al. (2014) in 268 patients with chronic neuropathic and musculoskeletal pain confirmed Cronbach’s alphas of 0.86 for Continuous, 0.82 for Intermittent, 0.84 for Neuropathic, 0.86 for Affective, and 0.93 for the Total score. Similarly, the French and German adaptations reported composite reliability indices and Cronbach’s alphas exceeding 0.85 across all dimensions.

Test-Retest Reliability

Temporal stability evaluated under clinically stable conditions over intervals ranging from 48 hours to 2 weeks has shown high reproducibility. Intraclass correlation coefficients (ICC) for the subscale and total scores consistently surpass acceptable psychometric benchmarks:

• Total Score: ICC = 0.88 to 0.94
• Continuous Pain: ICC = 0.84 to 0.89
• Intermittent Pain: ICC = 0.81 to 0.86
• Predominantly Neuropathic Pain: ICC = 0.87 to 0.92
• Affective Descriptors: ICC = 0.79 to 0.85

These metrics demonstrate that measurement error is minimal, ensuring that observed variations across repeated administrations reflect true therapeutic or natural changes in the underlying pain syndrome.

Factor Analysis

The dimensional architecture of the SF-MPQ-2 was rigorously established using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Exploratory Factor Analysis (EFA)

During scale construction, Dworkin et al. (2009) pooled pain descriptors from the original MPQ, revised clinical inventories, and expert panel reviews. Principal axis factoring with oblique promax rotation was applied to patient response datasets. The exploratory models clearly extracted four robust eigenvalues exceeding 1.0, explaining over 58% of the total variance. Items loaded cleanly onto their intended theoretical dimensions with primary factor loadings generally exceeding 0.55 and minimal cross-loadings (rarely exceeding 0.25).

Confirmatory Factor Analysis (CFA)

Subsequent CFA investigations evaluated competing structural models:

• Unidimensional Model: All 22 items loading onto a single general pain factor. (Demonstrated poor fit: χ²/df > 5.2, CFI < 0.80, RMSEA > 0.10).
• Two-Factor Model: Dividing items into classic sensory versus affective dimensions. (Demonstrated inadequate fit: CFI < 0.85, RMSEA > 0.085).
• Four-Factor Oblique Model: Continuous, Intermittent, Neuropathic, and Affective factors correlated freely. (Demonstrated optimal fit: Comparative Fit Index [CFI] = 0.952, Tucker-Lewis Index [TLI] = 0.945, Root Mean Square Error of Approximation [RMSEA] = 0.054 [90% CI: 0.048–0.060], and Standardized Root Mean Square Residual [SRMR] = 0.046).

Factor intercorrelations among the four latent factors typically range between r = 0.45 and r = 0.71, confirming that while these domains share underlying variance attributable to an overarching pain experience, they represent psychometrically distinct facets that warrant individual evaluation and profile tracking.

Instrument / Measurement Tool

The SF-MPQ-2 is a self-administered, paper-and-pencil or electronically administered questionnaire designed for rapid clinical and experimental execution.

• Test Type: Multi-dimensional patient-reported outcome measure (PROM).
• Target Population: Adult patients (ages 18+) presenting with acute, subacute, or chronic pain conditions, including musculoskeletal, inflammatory, neuropathic, and visceral pain syndromes. (Validated pediatric and adolescent adaptations also exist).
• Administration Time: Approximately 3 to 5 minutes to complete; under 2 minutes for automated electronic systems.
• Item Count: 22 standardized descriptors.
• Subscales:
  • Continuous Pain: Items 1, 5, 6, 8, 9, 10 (6 items)
  • Intermittent Pain: Items 2, 3, 4, 11, 16, 18 (6 items)
  • Predominantly Neuropathic Pain: Items 7, 17, 19, 20, 21, 22 (6 items)
  • Affective Descriptors: Items 12, 13, 14, 15 (4 items)
• Response Scale: 11-point Numerical Rating Scale (NRS) for each item, where:
  • 0 = none
  • 1 to 9 = intermediate gradations of symptom severity
  • 10 = worst possible
• Scoring Rules:
  • Subscale Mean Scores: Calculated by summing the item ratings within each subscale and dividing by the number of completed items in that subscale (yielding a continuous score ranging from 0.0 to 10.0).
  • Total Score: Calculated as the mean across all 22 items (sum of all items divided by 22), ranging from 0.0 to 10.0. Alternatively, an absolute total sum score (ranging from 0 to 220) may be reported if explicitly designated.
  • Missing Data: If more than 2 items in a 6-item subscale, or more than 1 item in the affective subscale, are missing, the respective subscale score is invalid. For mild missing data (e.g., 1 missing item), mean substitution of available items within that subscale is psychometrically acceptable.

Permissions & Fee and Test Year

The Short-Form McGill Pain Questionnaire (SF-MPQ) was initially developed in 1984 and published in 1987 by Ronald Melzack. The expanded and revised SF-MPQ-2 was published in 2009 by Robert H. Dworkin, Ronald Melzack, and the IMMPACT consensus team.

The copyright to the McGill Pain Questionnaire and its derivative versions is held by McGill University (Montreal, Canada) and Dr. Ronald Melzack’s estate, with licensing management overseen in part via commercial and academic distribution repositories. For non-funded academic research, individual clinical practice, and educational purposes, the instrument is generally accessible at no cost or for nominal processing fees upon formal registration through licensing entities such as Mapi Research Trust / PROQOLID (ePROVIDE). Commercial entities, industry-sponsored clinical trials, and proprietary healthcare software implementations require formal licensing agreements and payment of royalties prior to distribution or electronic integration.

References

• Dworkin, R. H., Turk, D. C., Revicki, D. A., Harding, G., Coyne, K. S., Peirce-Sandner, S., Bhagwat, D., Everton, D., Burke, L. B., Cowan, P., Farrar, J. T., Hertz, S., Max, M. B., Rappaport, B. A., & Melzack, R. (2009). Development and initial validation of an expanded and revised version of the Short-form McGill Pain Questionnaire (SF-MPQ-2). Pain, 144(1–2), 35–42. https://doi.org/10.1016/j.pain.2009.03.019
• Maruo, T., Nakae, A., Maeda, L., Shi, K., Takahashi, K., Morris, S., Hosomi, K., Kanatani, H., Matsuzaki, T., & Saitoh, Y. (2014). Validity, reliability, and assessment sensitivity of the Japanese version of the Short-Form McGill Pain Questionnaire 2 in Japanese patients with neuropathic and non-neuropathic pain. Pain Medicine, 15(11), 1930–1937. https://doi.org/10.1111/pme.12520
• Melzack, R. (1975). The McGill Pain Questionnaire: Major properties and scoring methods. Pain, 1(3), 277–299. https://doi.org/10.1016/0304-3959(75)90044-5
• Melzack, R. (1987). The short-form McGill Pain Questionnaire. Pain, 30(2), 191–197. https://doi.org/10.1016/0304-3959(87)91074-8
• Melzack, R. (1999). From the gate to the neuromatrix. Pain, 82(Suppl 1), S121–S126. https://doi.org/10.1016/S0304-3959(99)00145-1
• Melzack, R., & Torgerson, W. S. (1971). On the language of pain. Anesthesiology, 34(1), 50–59. https://doi.org/10.1097/00000542-197101000-00017
• Melzack, R., & Wall, P. D. (1965). Pain mechanisms: A new theory. Science, 150(3699), 971–979. https://doi.org/10.1126/science.150.3699.971

Items of the Scale