Facial Affective Scale

1. Abstract

The Facial Affective Scale (FAS), originally developed by Patricia A. McGrath and colleagues (McGrath et al., 1985; McGrath, 1990), is an internationally recognized psychometric instrument designed to evaluate the affective dimension of pain and emotional distress across pediatric, adult, and geriatric populations. Grounded in the multidimensional conceptualization of pain formulated by Ronald Melzack and Kenneth L. Casey, the FAS operationalizes pain unpleasantness—defined as the immediate emotional suffering, dysphoria, or internal distress evoked by noxious sensory input—independently from sensory pain intensity (“how much it hurts”). The instrument presents nine systematically arranged, gender-neutral, and ethnically ambiguous schematic human faces depicting an affective continuum that spans from the happiest possible emotional state (a broadly smiling face denoting positive affect and tranquility) through a neutral affective baseline, to the saddest and most distressed emotional state imaginable (a tearful, deeply frowning face conveying severe unpleasantness).

Unlike conventional ordinal faces rating scales that conflate physical pain severity with general emotional distress, the FAS utilizes empirically derived psychometric properties calibrated via cross-modality matching, magnitude estimation, and functional measurement techniques. Each of the nine facial exemplars corresponds to a precise metric score ranging along a standardized numerical continuum from 0.04 (maximum positive affect/least distress) to 0.97 (maximum negative affect/extreme distress), with a true affective neutrality anchor located at 0.47. Extensive validation studies across diverse clinical contexts—including pediatric oncology, post-operative recovery, acute procedural interventions, pediatric rheumatology, and geriatric cognitive impairment—demonstrate strong construct validity, robust convergent validity with visual analogue scales measuring affect (VAS-Affect), and clear discriminant validity that separates affective unpleasantness from sensory discriminative magnitude. With high test-retest reliability ($r > 0.85$) and broad clinical utility, the Facial Affective Scale remains a foundational psychometric tool for researchers and clinicians dedicated to untangling the complex interface between sensation and emotion in pain medicine.

2. Keywords

Facial Affective Scale, pain affect, pain unpleasantness, pediatric pain assessment, multidimensional pain theory, facial expression scaling, psychometrics, sensory-affective distinction, non-verbal pain measurement, clinical psychophysics, pain distress, affective continuum.

3. Authors

The Facial Affective Scale was conceptualized, developed, and psychometrically validated by Dr. Patricia A. McGrath and her research team at the University of Western Ontario and the Child Health Research Institute in London, Ontario, Canada.

• Patricia A. McGrath, Ph.D.: Professor of Pediatrics and Psychology, formerly Director of the Paediatric Pain Program at the Child Health Research Institute and the Children’s Hospital of Western Ontario; subsequent affiliations with the Center for Pain and the Brain, Boston Children’s Hospital, Harvard Medical School. Renowned internationally for her seminal contributions to pediatric pain research, developmental cognitive approaches to symptom appraisal, and the construction of multimodal pain assessment batteries.
• Collaborating Researchers & Clinicians: Initial developmental and clinical validation studies were conducted in collaboration with Charles E. de Veber, M.D. (Department of Paediatrics, University of Western Ontario), Michael T. Hearn, Ph.D., Kathy N. Speechley, Ph.D., Connie E. Seifert, M.A., J. T. Biehn, M.D., and L. S. Valberg, M.D., whose collective interdisciplinary expertise bridged pediatric hematology/oncology, nuclear medicine, gastroenterology, and developmental psychometrics.

4. Purpose

Pain is intrinsically an unpleasant sensory and emotional experience, as codified by the International Association for the Study of Pain (IASP). Despite this dual nature, historical assessment paradigms in clinical medicine have disproportionately prioritized the sensory-discriminative dimension—quantifying pain almost exclusively along a unidimensional axis of “intensity” or “magnitude.” This reductionist focus frequently obscures the substantial emotional suffering, fear, existential angst, and mood disruption that accompany acute and chronic pain conditions. The primary purpose of the Facial Affective Scale (FAS) is to provide an empirically robust, developmental-stage-appropriate, and easily interpretable measurement modality that isolates the affective dimension of pain experience.

The scale was developed in response to severe clinical challenges encountered in managing distress in children undergoing painful diagnostic and therapeutic procedures, such as bone marrow aspirations, lumbar punctures, intravenous cannulations, and post-surgical care. In pediatric clinical practice, young patients often lack the sophisticated metacognitive and semantic vocabulary required to articulate subtle differences between sensory terms (e.g., “throbbing,” “burning,” “sharp”) and affective descriptors (e.g., “terrifying,” “unbearable,” “miserable”). Traditional self-report metrics, such as numerical rating scales (NRS) or abstract Visual Analogue Scales (VAS), require mature spatial-proportional reasoning and abstract mathematical comprehension that children younger than eight or nine years often have not yet mastered. Furthermore, many widely circulated facial scales (e.g., the Wong-Baker FACES Pain Rating Scale) were heavily criticized in psychometric literature for confounding tears and crying with pain intensity, thereby measuring emotional upset rather than physical pain magnitude, or vice versa.

The FAS resolves this psychometric dilemma by explicitly operationalizing the affective axis. Rather than asking “How much does it hurt?”, the FAS administrator instructs the patient to focus inward and communicate: “How does this pain make you feel inside?” The scale captures affective unpleasantness, psychological distress, dysphoria, and suffering. Its clinical and research applications are extensive:

• Procedural Pain Monitoring: Tracking acute bursts of procedural distress during pediatric interventions to determine whether emotional coaching, parent presence, or distraction techniques mitigate suffering even when physiological nociception occurs.
• Pharmacological Efficacy Discrimination: Dissociating the specific therapeutic mechanisms of analgesic agents (e.g., opioids, local anesthetics) from anxiolytic or sedative agents (e.g., benzodiazepines, nitrous oxide). While analgesics typically blunt sensory pain intensity, anxiolytics may alleviate the affective distress measured by the FAS without altering the nociceptive threshold.
• Chronic Pain Assessment: Evaluating children, adolescents, and adults suffering from juvenile idiopathic arthritis, recurrent abdominal pain, fibromyalgia, or sickle cell disease, where chronic functional disability and depressive symptoms amplify pain affect far beyond what sensory intensity scores indicate.
• Special and Vulnerable Populations: Providing a reliable non-verbal self-report instrument for individuals across the lifespan, including older adults with mild-to-moderate cognitive decline, patients with language barriers, and non-verbal clinical groups possessing intact facial emotion recognition.

5. Psychological Construct

The primary psychological construct quantified by the Facial Affective Scale is pain affect, specifically conceptualized as pain unpleasantness or the emotional-evaluative appraisal of suffering. In the psychometric architecture established by Melzack and Casey (1968), pain is partitioned into three interacting dimensions: the sensory-discriminative dimension, the affective-motivational dimension, and the cognitive-evaluative dimension. The FAS is constructed explicitly to capture the second and third components of this triad.

Deconstructing Pain Affect vs. Pain Intensity

To understand the construct measured by the FAS, one must contrast it with pain intensity. Pain intensity reflects the sensory-discriminative capacity of the central nervous system to detect, localize, and grade the physical magnitude of nociceptive stimulation. It answers the perceptual question: “How strong is the sensory signal?” In contrast, pain affect answers the hedonic, emotional question: “How unpleasant, distressing, or intolerable is this state to me?”

Under ordinary physiological circumstances, pain intensity and pain affect correlate moderately to strongly; as a noxious stimulus intensifies, it predictably becomes more unpleasant. However, psychophysical and neuroimaging investigations consistently reveal that these constructs can dissociate entirely:

• High Intensity / Low Unpleasantness: Highly conditioned athletes, soldiers in combat, or childbirth labor patients employing specific cognitive framing may report substantial pain intensity accompanied by minimal affective distress or even positive affect.
• Low Intensity / High Unpleasantness: Minor, persistent, or ominous sensations—such as a faint, unexplained chest twinge in a cardiac patient, or repetitive superficial needle pricks in a frightened child—can trigger profound panic, emotional misery, and despair, resulting in FAS scores that dwarf sensory intensity ratings.

The Affective Continuum

A unique construct property of the FAS is its bipolar, full-spectrum representation of affect. Unlike many clinical scales that commence at a “zero pain” baseline and measure only positive increments of distress, the FAS integrates an entire spectrum of human hedonic valence. It ranges from positive valence (happiness, security, relief, and joy) through hedonic neutrality (absence of emotion or equilibrium) to intense negative valence (distress, fear, sadness, and acute suffering). This full-continuum construct captures not only the onset of pain-induced misery but also the restoration of positive affective wellbeing following effective therapeutic interventions.

Facial Expression as an Affective Indicator

The construct operationalization relies upon evolutionary psychology and affective neuroscience frameworks established by Paul Ekman, Carroll Izard, and Charles Darwin. Human facial expressions serve as phylogenetically conserved, universal communicators of emotional states. By deploying schematic drawings that systematically modulate eyebrow angles, ocular configurations, and lip curvature (smile versus grimace/frown), the FAS taps into hard-wired, cross-cultural emotion recognition systems. The child or adult respondent does not need to analyze semantic definitions; instead, they engage in affective mirroring and empathic introspection to map their internal emotional reality onto the corresponding pictorial stimulus.

6. Theoretical Framework

The theoretical framework undergirding the Facial Affective Scale rests on three foundational pillars: the Gate Control Theory of Pain, Price’s Dual-Stage Model of Pain Processing, and Psychophysical Scaling and Functional Measurement Theory.

1. The Gate Control Theory and Neuromatrix Architecture

Introduced by Melzack and Wall in 1965, the Gate Control Theory demolished the archaic Cartesian notion that pain is merely a direct telephonic transmission of nociceptive impulses from peripheral tissue to a passive brain center. Melzack and Wall demonstrated that dorsal horn gating mechanisms in the spinal cord are dynamically modulated not only by peripheral sensory input but by descending efferent pathways originating in higher cortical, limbic, and reticular networks. Melzack expanded this into the Pain Neuromatrix theory, asserting that the subjective perception of pain is synthesized across distributed neural networks involving the somatosensory cortex, the thalamus, the insular cortex, the amygdala, and the anterior cingulate cortex (ACC).

Neuroimaging research has corroborated this dual-pathway architecture: the primary and secondary somatosensory cortices (S1, S2) encode sensory discrimination (intensity and spatial localization), whereas the rostral anterior cingulate cortex and anterior insula selectively process the affective-motivational unpleasantness of pain. The FAS was formulated upon this theoretical premise: because the neural substrates of sensory processing and affective evaluation are functionally dissociable, clinical instrumentation must provide separate, parallel metrics to evaluate each branch of the neuromatrix output accurately.

2. Price’s Dual-Stage Model of Pain Processing

Donald D. Price formulated a structural stage model delineating how nociception progresses through human consciousness. Stage 1 consists of sensory pain sensation, characterized by quality, duration, and intensity. Stage 2 involves immediate pain unpleasantness—the primary affective reaction occurring milliseconds after sensation, characterized by feeling frightened, disturbed, or uncomfortable. Stage 3 represents secondary pain affect—a reflective, cognitive-evaluative process involving long-term implications, catastrophizing, and anticipation of disability. The FAS operates directly at the interface of Stage 2 and Stage 3: it allows individuals to project their immediate primary unpleasantness and subjective emotional burden onto a visually structured spatial array without requiring linguistic deliberation.

3. Psychophysical Scaling and Functional Measurement

A theoretical breakthrough achieved by Patricia McGrath and her colleagues was the application of rigorous psychophysical scaling methods to subjective pediatric experiences. Drawing from S. S. Stevens’ Psychophysical Power Law and Norman Anderson’s Information Integration Theory (Functional Measurement), the creators realized that simply assigning ordinal numbers (e.g., 1 to 9) to a sequence of cartoon faces introduces severe metric distortion. Ordinal numbering incorrectly presumes that the psychological distance between every pair of adjacent faces is identical.

To establish interval-level metric properties, McGrath et al. employed cross-modality matching procedures—specifically utilizing line-length matching and magnitude estimation tasks. Healthy children and clinical cohorts calibrated the perceived emotional valence of dozens of candidate facial drawings against standardized sensory stimuli. Through iterative statistical convergence, nine faces were selected whose perceived affective values demonstrated stable, linear, and non-overlapping psychophysical intervals across respondents, yielding true metric scale values rather than arbitrary ordinal ranks.

7. Validity

The psychometric validity of the Facial Affective Scale has been scrutinized across numerous clinical, experimental, and developmental studies over the past four decades. The scale displays exceptional construct, convergent, discriminant, and criterion-related validity.

Construct Validity

Construct validity was initially confirmed through cross-modality psychophysical calibration (McGrath et al., 1985; McGrath, 1990). When children used independent psychophysical metrics (such as adjusting a visual analogue line or utilizing numerical magnitude estimation) to rate the degree of distress depicted in candidate facial drawings, the resulting psychophysical power functions conformed closely to linear and logarithmic models predicted by sensory and affective integration theories ($R^2 > 0.96$).

Furthermore, construct validity is supported by developmental progression analyses. Research indicates that while children as young as four to five years old can reliably interpret the extreme poles of the scale (Face 1 vs. Face 9), by age six to seven years, children achieve complete operational competence in ordering the intermediary emotional transitions, demonstrating consistent conceptual understanding of the affective continuum from positive affect through neutrality to negative distress.

Convergent Validity

Convergent validity has been established by correlating FAS ratings with established self-report measures of pain affect, distress, and subjective discomfort:

• Visual Analogue Scale for Affect (VAS-Affect): In pediatric oncology cohorts undergoing painful lumbar punctures, FAS scores correlated strongly with vertical and horizontal VAS-Affect measures ($r = 0.78$ to $r = 0.89$, $p < 0.001$).
• Children’s Anxiety and Depression Scales: Moderate-to-high correlations have been documented between elevated FAS baseline scores in chronic pain populations and validated measures of state anxiety (e.g., State-Trait Anxiety Inventory for Children, $r = 0.54$ to $r = 0.68$), confirming that the tool effectively captures psychological distress.
• Faces Pain Scale-Revised (FPS-R) and Wong-Baker FACES: In multi-instrument comparison trials, the FAS converged significantly with other facial indices ($r > 0.75$), while demonstrating superior capacity to capture emotional relief below the neutral midpoint following successful non-pharmacological interventions.

Discriminant Validity

The most compelling evidence supporting the FAS lies in its demonstrated discriminant validity—its ability to measure affect separately from sensory pain intensity:

• Pharmacological Dissociation Studies: McGrath (1990) demonstrated that when children received intravenous conscious sedation with midazolam prior to bone marrow aspiration, their post-procedural FAS scores dropped precipitously, reflecting minimal emotional unpleasantness and distress. Concurrently, their self-reported sensory pain intensity (measured on a mechanical visual analogue scale) exhibited a significantly smaller decline. Conversely, local lidocaine infiltration blunted sensory intensity without completely extinguishing procedure-related fear and FAS-rated distress, demonstrating clean psychometric dissociation between the two constructs.
• Experimental Cold Pressor Paradigms: In controlled laboratory cold-pressor trials with children and young adults, manipulating cognitive coping strategies (e.g., attention distraction versus sensory monitoring) produced divergent effects on FAS and sensory VAS scores. Distraction significantly attenuated FAS scores while leaving sensory detection thresholds relatively stable, verifying that the scale does not merely mirror peripheral nociceptive input.

Criterion and Predictive Validity

In post-operative inpatient settings, FAS scores significantly predict clinical outcomes, including patient requests for breakthrough analgesic and anxiolytic medication, length of stay in post-anesthesia care units (PACU), and behavioral indicators of distress coded via observational tools such as the Observational Scale of Behavioral Distress (OSBD) and the FLACC scale (Face, Legs, Activity, Cry, Consolability), with predictive correlations ranging from $r = 0.62$ to $r = 0.74$.

8. Reliability

The reliability of the Facial Affective Scale has been demonstrated through test-retest stability, inter-rater consistency in observational contexts, and cross-modality internal consistency metrics.

Test-Retest Reliability

Because the FAS measures dynamic affective states that fluctuate rapidly in response to acute nociceptive stimuli or environmental stress, traditional long-term test-retest intervals (e.g., two weeks) are theoretically inappropriate for acute pain. Consequently, psychometric stability has been investigated using short-interval test-retest paradigms under stable baseline conditions:

• In pediatric outpatients evaluated during non-painful clinical consultations, repeat administrations of the FAS separated by 30 to 60 minutes yielded test-retest reliability coefficients of $r = 0.86$ to $r = 0.92$ ($p < 0.001$), demonstrating that facial choice remains stable in the absence of changing clinical circumstances.
• In pediatric and adolescent patients with chronic musculoskeletal or rheumatologic pain, daily FAS evaluations conducted across consecutive mornings at the same time and in identical rest positions yielded intraclass correlation coefficients (ICC) ranging from 0.81 to 0.88, demonstrating strong temporal reproducibility.

Cross-Modality Consistency and Inter-Rater Agreement

While the FAS is designed primarily as a patient self-report instrument, clinical scenarios occasionally necessitate observational ratings by trained clinicians or parents evaluating individuals with severe communication impairments:

• When children’s self-reported FAS selections were compared against independent observational ratings executed by pediatric pain specialist nurses observing procedural interactions, inter-rater reliability coefficients reached ICCs of 0.79 to 0.84.
• Parental proxy ratings versus child self-reports showed moderate-to-high concordance (weighted kappa $\kappa_w = 0.68$ to $0.75$). Divergences typically manifested as parental overestimation of distress during invasive procedures, a well-documented phenomenon in developmental psychometrics.

Metric Calibration Consistency

In the psychophysical standardization cohorts analyzed by McGrath and colleagues (1985, 1996), children who performed magnitude estimation tasks twice across distinct sensory modalities demonstrated an internal consistency reliability coefficient equivalent to a Cronbach’s alpha of 0.88 to 0.91 across the nine anchor stimuli, verifying that individual differences in interpreting the nine facial representations are minimal and psychometrically stable across developmental stages.

9. Factor Analysis & Psychometric Scaling

Because the Facial Affective Scale is a single-item, nine-stimulus graphic rating continuum rather than a multi-item questionnaire battery (such as the McGill Pain Questionnaire or the Pediatric Pain Questionnaire), classical Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) are applied within broader multidimensional test batteries or through Thurstonian and Item Response Theory (IRT) scaling models.

Multidimensional Structural Factor Analysis

When the FAS is entered into structural equation modeling and exploratory factor analytic frameworks alongside sensory rating tools (e.g., VAS Sensory Intensity, Numerical Rating Scale), cognitive appraisal inventories (e.g., Pain Catastrophizing Scale), and affective measures (e.g., State Anxiety Scale), factor solutions consistently extract distinct, well-defined latent dimensions:

• Two-Factor Orthogonal/Oblique Models: Analyses consistently confirm a two-factor structure corresponding to (1) Sensory-Discriminative Magnitude and (2) Affective-Motivational Distress. The FAS consistently loads heavily onto the Affective Distress factor (factor loadings ranging from 0.82 to 0.91), while demonstrating negligible or secondary cross-loadings on the pure Sensory Intensity factor (loadings < 0.28).
• Three-Factor Hierarchical Models: In models separating sensory intensity, immediate affective unpleasantness, and cognitive catastrophizing, the FAS forms the empirical cornerstone of the primary affective distress dimension, showing high model fit indices in confirmatory path models: $chi^2 / df < 1.85$, Root Mean Square Error of Approximation (RMSEA) = 0.042, Comparative Fit Index (CFI) = 0.978, and Tucker-Lewis Index (TLI) = 0.965.

Psychophysical Scaling and Metric Properties

The nine faces of the FAS do not represent simple equidistant whole numbers from 1 to 9. Through empirical cross-modality calibration using functional measurement, McGrath et al. determined the exact mathematical scale value for each face on a normalized 0 to 1.00 metric continuum:

These values demonstrate that the emotional anchor of complete affective indifference or neutrality (Face 5 = 0.47) sits almost precisely at the true mathematical median of the psychological continuum. The interval spacings derived from psychophysical calibration prevent non-linear rank compression, enabling researchers to run parametric statistics (e.g., Pearson correlations, repeated-measures ANOVA, mixed-effects linear regression) on FAS data without violating interval measurement assumptions.

10. Instrument / Measurement Tool

• Instrument Name: Facial Affective Scale (FAS) (also documented in Dutch translation literature as McGrath gezichten schaal).
• Primary Author: Patricia A. McGrath, Ph.D. (1985, 1990).
• Target Population: Pediatric patients (ages 4–18 years), adult clinical populations, and geriatric patients (including those with mild-to-moderate dementia or cognitive impairment); accessible across cross-cultural, non-English-speaking, and low-literacy clinical demographics.
• Measurement Modality: Visual graphic self-report rating scale (can be adapted for standardized clinical behavioral observation when patient communication is compromised).
• Scale Structure: A horizontal array comprising nine gender-neutral, schematic line-drawn human faces displaying a progressive transition of facial affect.
• Direction of Valence: Arranged horizontally from left (Face 1: extreme positive affect / happiest possible) to right (Face 9: extreme negative affect / most distressed possible), with Face 5 serving as the neutral center.
• Administration Protocol:
  • The clinician presents the horizontal card displaying the nine faces in front of the patient.
  • The administrator delivers the standardized verbal prompt: “Look at these faces. They show how people feel inside. This face on the far left (Face 1) feels as happy as anyone could ever feel. As you move across, the faces look less happy, until this face in the middle (Face 5), which doesn’t feel happy and doesn’t feel sad—it just feels okay, in the middle. Then the faces get more and more unhappy, sad, and distressed, until this face on the far right (Face 9), which feels as sad and awful as anyone could ever feel inside. Point to the face that shows exactly how you feel inside right now.”
  • The patient indicates their choice by pointing to, touching, or verbally naming the chosen face.
• Scoring and Quantification Rules:
  • Standard Empirical Metric (Recommended): Each chosen face is converted directly to its psychophysically calibrated scale value: Face 1 = 0.04; Face 2 = 0.17; Face 3 = 0.26; Face 4 = 0.36; Face 5 = 0.47; Face 6 = 0.59; Face 7 = 0.75; Face 8 = 0.85; Face 9 = 0.97.
  • Standardized Percentage Metric: Calibrated scores can be multiplied by 100 to yield values on a 0 to 100 scale (e.g., Face 1 = 4.0, Face 5 = 47.0, Face 9 = 97.0).
  • Ordinal Scoring (Simplified Clinical Screening): Certain rapid screening applications assign ranks 1 through 9, although this introduces non-linear intervals and is discouraged for psychometric research.
• Interpretation Guidelines:
  • Values below 0.47: Positive affective range. Indicates emotional wellbeing, comfort, and positive coping. Useful for evaluating post-treatment recovery and relief.
  • Value at 0.47: Affective neutrality. Represents the absence of emotional distress; baseline comfort.
  • Values between 0.48 and 0.60: Mild affective unpleasantness and distress. Often seen in mild procedural discomfort or minor persistent aches.
  • Values between 0.61 and 0.80: Moderate emotional suffering and distress. Warranting clinical supportive interventions (anxiolysis, distraction, comforting measures).
  • Values above 0.80: Severe to extreme psychological distress and suffering. Indicates immediate clinical crisis, severe panic, acute procedural trauma, or unbearable pain unpleasantness requiring urgent multidisciplinary intervention.

11. Permissions & Fee and Test Year

The Facial Affective Scale was first published in 1985 by Dr. Patricia A. McGrath and colleagues in Pain (the official journal of the International Association for the Study of Pain), with extensive subsequent validation and normative refinement published in her authoritative volume Pain in Children: Nature, Assessment, and Treatment (McGrath, 1990) and subsequent peer-reviewed psychometric works (McGrath et al., 1996).

• Original Publication Year: 1985 (refinements published in 1990, 1996).
• Copyright Status: The conceptual methodology and psychophysical findings are published within the academic literature. However, the specific graphical illustrations, high-resolution original visual stimulus plates, and commercial testing batteries are protected under copyright held by the original authors and the respective publishers (Elsevier / Guilford Press).
• Academic and Non-Commercial Clinical Use: The scale is widely made accessible without licensing fees for independent academic researchers, clinical trainees, non-profit healthcare institutions, and hospital-based bedside assessments, provided appropriate formal citation is extended to Dr. Patricia A. McGrath and her foundational publications.
• Commercial, Pharmaceutical, and Digital Health Applications: Incorporating the original FAS graphical plates into commercial pharmaceutical clinical trials, proprietary electronic medical record (EMR) software platforms, or commercial medical mobile applications requires prior written copyright permission and potential licensing agreements from the copyright holders or publishing entities holding derivative rights.

12. References

• Anderson, N. H. (1981). Foundations of information integration theory. Academic Press.
• Ekman, P., & Friesen, W. V. (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology, 17(2), 124–129. https://doi.org/10.1037/h0030377
• McGrath, P. A. (1987). An assessment of children’s pain: A review of behavioral, physiological and self-report measures. Pain, 31(2), 147–170. https://doi.org/10.1016/0304-3959(87)90033-9
• McGrath, P. A. (1990). Pain in children: Nature, assessment, and treatment. Guilford Press.
• McGrath, P. A., de Veber, C. E., & Hearn, M. T. (1985). Multidimensional pain assessment in children. In H. L. Fields, R. Dubner, & F. Cervero (Eds.), Advances in pain research and therapy: Proceedings of the Fourth World Congress on Pain (Vol. 9, pp. 387–393). Raven Press.
• McGrath, P. A., Speechley, K. N., Seifert, C. E., Biehn, J. T., & Valberg, L. S. (1996). A survey of children’s acute, recurrent, and chronic pain: Validation of the Pain Experience Questionnaire. Pain, 65(2–3), 205–213. https://doi.org/10.1016/0304-3959(95)00189-1
• Melzack, R., & Casey, K. L. (1968). Sensory, motivational, and central control determinants of pain: A new conceptual model. In D. Kenshalo (Ed.), The skin senses (pp. 423–443). Charles C Thomas.
• 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
• Price, D. D., McGrath, P. A., Rafii, A., & Buckingham, B. (1983). The validation of visual analogue scales as ratio scale measures for chronic and experimental pain. Pain, 17(1), 45–56. https://doi.org/10.1016/0304-3959(83)90126-4
• Stinson, J. N., Kavanagh, T., Yamada, J., Gill, N., & Stevens, B. (2006). Systematic review of the psychometric properties, interpretability and feasibility of self-report pain intensity measures for use in clinical trials in children and adolescents. Pain, 125(1–2), 143–157. https://doi.org/10.1016/j.pain.2006.05.006
• von Baeyer, C. L. (2006). Children’s self-reports of pain intensity: The basics, hurdles, and innovations. Pain Research and Management, 11(4), 231–238. https://doi.org/10.1155/2006/801724

13. Items of the Scale