Comfort Scale

1. Abstract

The COMFORT Scale is a multidimensional clinical assessment instrument designed to measure distress, sedation adequacy, and postoperative pain in critically ill pediatric patients, ranging from neonates to adolescents under 18 years of age. Developed originally by Ambuel et al. (1992) to address the clinical challenge of quantifying pain and agitation in non-verbal or mechanically ventilated children within the Pediatric Intensive Care Unit (PICU), the scale integrates both behavioral observations and physiological parameters. The classical instrument comprises eight discrete items: six behavioral dimensions (Alertness, Calmness/Agitation, Respiratory Response, Physical Movement, Muscle Tone, and Facial Tension) and two physiological dimensions (Baseline Heart Rate and Baseline Mean Arterial Blood Pressure). Each dimension is operationalized across an anchored 5-point rating scale ranging from 1 to 5, yielding a cumulative total score between 8 and 40, wherein scores between 17 and 26 traditionally demarcate optimal comfort and sedation.

Extensive psychometric investigations have established robust psychometric properties across diverse clinical populations. Internal consistency reliability typically achieves Cronbach’s α coefficients exceeding .84 to .90 across general intensive care cohorts, with inter-rater reliability intraclass correlation coefficients (ICC) ranging from .84 to .93 among trained clinical nurses. Structural validity evaluations through exploratory and confirmatory factor analyses consistently elucidate a dominant behavioral factor, which later prompted the development of the modified behavioral version, the COMFORT-B Scale by van Dijk et al. (2000). The COMFORT Scale demonstrates excellent convergent validity when correlated against visual analog scales of pain and distress (Pearson’s r ≈ .68 to .84) and discriminant validity across escalating analgesic and sedative dosing tiers. The instrument serves as an indispensable gold standard in neonatal and pediatric critical care research and clinical protocol titration.

2. Keywords

COMFORT scale, pediatric intensive care unit, pain assessment, behavioral distress, sedation monitoring, mechanical ventilation, pediatric psychometrics, neonatal pain, COMFORT-B, physiological markers, clinical observation, analgesia titration

3. Authors

The original COMFORT Scale was conceptualized, developed, and validated by an interdisciplinary team led by Bruce Ambuel, Ph.D., Department of Family and Community Medicine, Medical College of Wisconsin; Kathleen W. Hamlett, Ph.D.; Ronald S. Marx, M.D.; and Joseph L. Blumer, Ph.D., M.D., affiliated with the Department of Pediatrics and Division of Pediatric Pharmacology and Critical Care at Rainbow Babies and Children’s Hospital, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA.

Subsequent psychometric refinement and cultural adaptation of the behavioral derivative (COMFORT-behavior or COMFORT-B) were executed by Monique van Dijk, Ph.D., along with Dick Tibboel, M.D., Ph.D., and colleagues within the Department of Pediatric Surgery and Intensive Care at the Sophia Children’s Hospital, Erasmus University Medical Center, Rotterdam, The Netherlands. Academic correspondence regarding original implementation methodologies historically directs through the pediatric critical care and behavioral medicine research units at Rainbow Babies & Children’s Hospital and the Erasmus MC Sophia Children’s Hospital registry.

4. Purpose

Assessing pain, agitation, and psychological distress in critically ill pediatric populations presents an immense clinical and psychometric hurdle. Due to anatomical development, cognitive immaturity, pervasive neurological illness, and the routine administration of endotracheal intubation accompanied by neuromuscular blockade or continuous intravenous sedation, traditional self-report mechanisms—the unequivocal gold standard in adult pain psychometrics—are largely untenable. The primary purpose of the COMFORT Scale is to furnish a standardized, objective, and reproducible multidimensional observational rubric that quantifies distress intensity and frequency in neonatal, infant, and pediatric patients undergoing intensive critical care management.

From an applied clinical perspective, the scale functions as an indispensable monitoring metric for guiding pharmacotherapeutic titration. Critically ill children admitted to PICU environments often suffer from acute physiological destabilization caused by invasive supportive technologies (e.g., positive pressure mechanical ventilation, arterial cannulation, and thoracostomy tubes). Under-sedation risks severe adverse events, including accidental extubation, displacement of vascular access catheters, severe hypoxemia secondary to dyssynchrony with mechanical ventilators, and excessive neuroendocrine stress responses. Conversely, over-sedation significantly prolongs the duration of mechanical ventilation, exacerbates iatrogenic physical dependency, precipitates withdrawal syndromes upon taper, and increases susceptibility to secondary nosocomial infections. The COMFORT Scale provides a calibrated baseline allowing critical care practitioners to navigate this tight therapeutic window effectively.

In academic and clinical research contexts, the COMFORT Scale serves as a primary endpoint in pharmacological trials evaluating novel sedative agents (such as dexmedetomidine, remifentanil, or propofol) and non-pharmacological soothing strategies. The scale operates within a theoretical framework recognizing that pain and distress manifest through simultaneous somatic, motoric, neurovegetative, and affective channels. By combining observable behavioral repertoires with objective hemodynamics, the tool minimizes observer bias and furnishes empirical rigor to environments where communication is radically impaired.

5. Psychological Construct

The COMFORT Scale operationalizes distress as a multifaceted psychological and somatic construct comprising behavioral arousal, motor agitation, muscular tension, and autonomic hyper-reactivity. The instrument is partitioned into eight discrete items reflecting these behavioral and physiological domains:

Behavioral Dimensions

• Alertness: Evaluates the patient’s baseline level of consciousness and interaction with the immediate sensory environment. A fully relaxed infant may demonstrate a score of 1 (deeply asleep with eyes closed and no spontaneous reaction to subtle acoustic stimuli), whereas an agitated child facing respiratory compromise might display a score of 5 (hyperalert, displaying wide and frightened eyes, hypersensitivity to non-noxious environmental perturbations).
• Calmness / Agitation: Measures the affective state and internal distress level of the patient. The continuum tracks the emotional transition from serenity (calm and peaceful) to severe autonomic-emotional perturbation, culminating in panicky, uncontrolled emotional distress (score of 5), where the child appears inconsolable and terror-stricken.
• Respiratory Response: Captures the interaction between a mechanically ventilated patient’s endogenous respiratory drive and the mechanical ventilator circuit. Ratings range from complete synchronization without cough (score of 1) to active ventilator asynchrony, vigorous coughing, biting the endotracheal tube, or frank fighting against ventilator insufflation (score of 5).
• Physical Movement: Quantifies the frequency, velocity, and spatial distribution of somatic motor activity. It differentiates between utter immobility (score of 1) and wild, uncoordinated, vigorous flailing involving the extremities, head, and trunk (score of 5), which jeopardizes therapeutic lines and patient safety.
• Muscle Tone: Assessed via passive flexion and extension of the major peripheral extremities. It anchors at total flaccidity (score of 1, often observed under deep anesthesia or excessive neuromuscular blockade), progressing through normal, euthermic resistance (score of 3), to rigid muscular hypertonicity characterized by tight finger/toe flexion and pervasive somatic spasticity (score of 5).
• Facial Tension: Evaluates the micro- and macro-expressions of the facial musculature, specifically the contraction of the brow, orbit, and nasolabial folds. Completely relaxed facial features (score of 1) contrast sharply with sustained grimacing, brow furrowing, and distorted, contorted facial mimicry (score of 5), indicative of acute nociception.

Physiological Dimensions

• Blood Pressure (MAP) Baseline: Quantifies autonomic sympathetic nervous system output reflected in Mean Arterial Pressure (MAP) relative to patient-specific resting baselines. Ratings reflect deviations from baseline, moving from sub-baseline values (score of 1) up to sustained elevations of over 15% above resting baseline (score of 5).
• Heart Rate Baseline: Acts as an index of sympathovagal balance, assessing tachycardia secondary to acute nociceptive signaling. Ratings categorize hemodynamic surges, scoring an elevated resting heart rate sustained at greater than 15% above the baseline as an indicator of physiological stress (score of 5).

6. Theoretical Framework

The theoretical architecture of the COMFORT Scale is anchored in the biobehavioral model of pain and stress, synthesizing concepts from Hans Selye’s General Adaptation Syndrome, modern neurobiology of pain, and developmental behavioral psychometrics. Pain and distress are not viewed merely as sensory reflexes, but rather as dynamic neurobiological events involving reciprocal interactions between the limbic system, the hypothalamic-pituitary-adrenal (HPA) axis, and the autonomic nervous system.

When nociceptive or psychological threats impact a vulnerable child, ascending spinothalamic tracts transmit impulses to thalamocortical networks while simultaneously activating the reticular activating system and periaqueductal gray. This produces a dual-pronged response: immediate, observable behavioral alterations (facial grimacing, motor restlessness, and defensive postures) aimed at evading the noxious stimulus, and involuntary autonomic mobilization (elevated catecholamine release, tachycardia, systemic vasoconstriction). The COMFORT Scale posited that measuring both expressions concurrently would provide a more valid estimate of the internal state than evaluating either domain in isolation.

However, developmental psychometrics highlights the challenge of physiological reactivity in critically ill infants. As later demonstrated by van Dijk et al. (2000), physiological parameters such as heart rate and arterial pressure are inherently subject to confounding critical care phenomena, including sepsis, volume depletion, inotropic infusions, and intrinsic cardiac pathology. Consequently, modern theoretical iterations (e.g., the COMFORT-B model) re-weight behavioral parameters as the more specific and sensitive indicators of central pain processing, whereas physiological items reflect systemic stress reactivity that must be interpreted alongside behavioral signs.

7. Validity

The COMFORT Scale has undergone extensive international validation, establishing rigorous construct, criterion, convergent, and discriminant validity across broad pediatric populations:

Construct and Criterion Validity

In the seminal validation trial by Ambuel et al. (1992), construct validity was established by comparing COMFORT scores against parallel global ratings provided by experienced intensive care nurses and pediatric psychologists. COMFORT scores correlated significantly with concurrent nurse visual analog ratings of distress (r = .75, p < .001). Criterion validity has been demonstrated by evaluating the scale’s capacity to identify clinically validated levels of inadequate sedation or pain, showing receiver operating characteristic (ROC) area under the curve (AUC) metrics routinely falling between .86 and .94.

Convergent and Discriminant Validity

Convergent validity has been repeatedly corroborated against established single-dimension and multidimensional instruments. Strong correlations are documented between the behavioral items of the COMFORT Scale and the Face, Legs, Activity, Cry, Consolability (FLACC) scale (Spearman’s ρ ranging from .73 to .89). Conversely, discriminant validity has been evidenced by the instrument’s ability to cleanly differentiate between distinct clinical interventions: scores drop precipitously following targeted opioid or benzodiazepine boluses (mean drops of 6 to 12 points post-intervention, p < .001). Furthermore, discriminant analyses distinguish pain-driven distress from mere hemodynamic volatility driven by pyrexia or hypovolemia, particularly when the behavioral sub-indices are analyzed.

8. Reliability

The psychometric reliability of the COMFORT Scale is supported by deep empirical literature documenting high internal consistency and exceptional inter-rater agreement across bedside clinicians.

Internal Consistency

Original reports by Ambuel et al. demonstrated a global Cronbach’s α of .84. Subsequent multicenter studies validating pediatric postoperative cohorts documented alpha coefficients spanning .84 to .90 across ventilated populations. When the behavioral components are isolated (as formalized in the COMFORT-B), Cronbach’s α often increases further to .87–.93, indicating that the elimination of labile autonomic parameters reduces measurement noise and enhances internal item covariance.

Inter-Rater Reliability

Given the observational nature of the COMFORT Scale, inter-observer agreement is critical. Inter-rater reliability evaluated using Intraclass Correlation Coefficients (ICC) and Cohen’s kappa across trained observers yields ICC values consistently between .84 and .93. Weighted kappa coefficients (κ) for individual items range from .63 (Facial Tension and Muscle Tone, which carry minor subjective ambiguity) to over .88 (Alertness, Calmness/Agitation, and Baseline Heart Rate). Test-retest reliability across brief, clinically stable intervals (10 to 15 minutes without change in sedation) shows stability coefficients of r > .80.

9. Factor Analysis

Structural evaluations through Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have yielded critical insights into the internal architecture of the COMFORT Scale.

Exploratory Factor Analysis

Early factor analyses by Ambuel et al. and subsequent structural re-analyses by van Dijk et al. (2000) revealed that the 8 items do not conform to a purely single unifactorial structure. Instead, principal axis factoring with varimax and oblimin rotations typically uncovers two robust latent factors:

• Factor 1: Behavioral Expression of Distress. Comprising Alertness, Calmness/Agitation, Respiratory Response, Physical Movement, Muscle Tone, and Facial Tension. This behavioral factor accounts for the vast majority of the variance (typically 48% to 58% of explained variance), with individual item factor loadings consistently ranging from .65 to .88.
• Factor 2: Autonomic / Physiological Reactivity. Comprising Baseline Blood Pressure and Baseline Heart Rate. This second factor accounts for roughly 12% to 16% of total variance, with factor loadings between .60 and .79. Notably, cross-loadings between physiological items and the behavioral items are remarkably weak (< .20), evidencing theoretical and statistical divergence between behavioral displays of distress and sympathetic tone in critically ill cohorts.

Confirmatory Factor Analysis & Goodness-of-Fit

Confirmatory factor analytic models evaluating the traditional 8-item two-factor solution demonstrate satisfactory model fit indices across pediatric intensive care datasets:

• Comparative Fit Index (CFI): .94 to .97
• Tucker-Lewis Index (TLI): .92 to .95
• Root Mean Square Error of Approximation (RMSEA): .051 to .068 (90% CI: .038–.078)
• Standardized Root Mean Square Residual (SRMR): .042 to .055

Because the physiological factor accounts for substantially less variance and is easily biased by vasoactive medications, van Dijk and colleagues validated the pure behavioral 6-item model (COMFORT-B), which exhibits even superior single-factor unidimensionality in CFA (CFI > .98, RMSEA < .05).

10. Instrument / Measurement Tool

The COMFORT Scale is structured as an observational clinical rating tool administered by healthcare personnel (nurses, physicians, clinical researchers) following a disciplined observation protocol.

• Test Type: Multidimensional Clinical Observation Rating Scale (combining systematic behavioral observation and physiological monitoring).
• Target Population: Neonates, infants, young children, and adolescents up to 18 years of age in pediatric intensive care, post-surgical, or critical care settings.
• Administration Format: Bedside observation. The rater directly observes the undisturbed patient for exactly two consecutive minutes, evaluating spontaneous alertness, movement, and facial expressions. The rater then assesses passive muscle tone via gentle physical elbow/knee flexion, and cross-checks the current Mean Arterial Blood Pressure and Heart Rate against pre-established baseline parameters.
• Item Count: 8 items (6 behavioral items and 2 physiological items).
• Response Scale: 5-point rating scale (1 to 5) for each dimension with specific operational anchors.
• Scoring and Interpretation Rules:
  • Each item is assigned an integer value from 1 to 5 based on explicit criteria.
  • Cumulative total scores range from a minimum of 8 to a maximum of 40.
  • Total Score 8 – 16: Over-sedation / Deep sedation (patient displays minimal to no reactivity; may risk therapeutic over-dosage).
  • Total Score 17 – 26: Optimal sedation and comfort (adequate pain control, calmness, and physiological stability).
  • Total Score 27 – 40: Inadequate sedation, acute agitation, pain, or distress (warrants immediate clinical evaluation, analgesic/sedative administration, or environmental stabilization).

11. Permissions & Fee and Test Year

The COMFORT Scale was originally formulated and published in 1992 by Bruce Ambuel and colleagues in the journal Critical Care Medicine. The adapted Dutch behavioral version (COMFORT-B) was subsequent published in 2000 by Monique van Dijk and colleagues in Intensive Care Medicine.

In terms of copyright and clinical utilization, the scale is widely accessible in the public academic domain for non-profit clinical, educational, and research implementations. Commercial software integrations, proprietary electronic health record (EHR) embedments, or fee-charging pharmaceutical trials frequently require formal academic citation and, where applicable, administrative permission from the original copyright holders (Lippincott Williams & Wilkins / Wolters Kluwer Health for the 1992 publication; Springer Nature for the 2000 publication; or the Erasmus MC Sophia Children’s Hospital intellectual property framework). There is typically no license fee for individual researchers or hospital systems conducting bedside care and investigator-driven trials.

12. References

Below are primary foundational references detailing the development, psychometric properties, and clinical application of the COMFORT and COMFORT-B scales:

• Ambuel, B., Hamlett, K. W., Marx, C. M., & Blumer, J. L. (1992). Assessing distress in pediatric intensive care environments: The COMFORT scale. Critical Care Medicine, 20(7), 957–964. https://doi.org/10.1097/00003246-199207000-00014
• van Dijk, M., de Boer, J. B., Koot, H. M., Tibboel, D., Passchier, J., & Duivenvoorden, H. J. (2000). The reliability and validity of the COMFORT scale as a postoperative pain instrument in 0 to 3-year-old infants. Pain, 84(2-3), 367–377. https://doi.org/10.1016/s0304-3959(99)00239-0
• van Dijk, M., Peters, J. W., van Deventer, P., & Tibboel, D. (2005). The COMFORT Behavior Scale: A tool for assessing pain and sedation in infants. American Journal of Nursing, 105(1), 33–36. https://doi.org/10.1097/00000446-200501000-00019
• Ista, E., van Dijk, M., Tibboel, D., & de Hoog, M. (2005). Assessment of sedation levels in pediatric intensive care patients can be improved by using the COMFORT ‘behavior’ scale. Pediatric Critical Care Medicine, 6(1), 58–63. https://doi.org/10.1097/01.PCC.0000149318.47417.A7
• Marx, C. M., Smith, P. G., Lowrie, L. H., Hamlett, K. W., Ambuel, B., Yamashita, T. S., & Blumer, J. L. (1994). Optimal sedation of mechanically ventilated pediatric critical care patients: A prospective evaluation of continuous midazolam infusion. Critical Care Medicine, 22(1), 163–170. https://doi.org/10.1097/00003246-199401000-00028
• Boerlage, A. A., Ista, E., Duivenvoorden, H. J., de Wildt, S. N., Tibboel, D., & van Dijk, M. (2015). The COMFORT behaviour scale detects clinically meaningful changes in neonates and infants with pain. European Journal of Pain, 19(4), 473–479. https://doi.org/10.1002/ejp.569

13. Items of the Scale