Neonatal Pain and Discomfort Scale

1. Abstract

The Neonatal Pain and Discomfort Scale (most widely known by its original French title, Échelle Douleur Inconfort Nouveau-né [EDIN]) is an established, clinician-administered observational instrument designed to quantify prolonged, persistent, or chronic pain and discomfort in preterm and full-term hospitalized neonates. Developed in response to the historical under-recognition of persistent neonatal distress within neonatal intensive care units (NICU), the EDIN diverges fundamentally from acute procedural pain tools by measuring behavioral patterns accumulated across an extended observation window (typically spanning several hours to a full nursing shift). The scale is unidimensional and comprises five operationalized behavioral indicators: facial expression, body movements, quality of sleep, quality of contact with caregivers, and consolability. Each indicator is scored along a four-point ordinal Likert scale ranging from 0 to 3, yielding a cumulative composite score between 0 and 15, where scores strictly greater than 5 indicate clinically actionable prolonged pain demanding immediate multimodal or pharmacological intervention.

Extensive psychometric investigations have established the scale’s robust statistical viability across diverse neonatal patient demographics. The EDIN demonstrates exemplary internal consistency, with Cronbach’s alpha coefficients consistently reported between .86 and .92 across both validation cohorts and subsequent cross-cultural translation trials. Inter-rater reliability demonstrates strong clinician concordance, yielding intraclass correlation coefficients (ICC) and weighted Cohen’s kappa coefficients ranging from .70 to .89 across independent nursing and physician dyads. Construct and convergent validity are evidenced by marked reductions in EDIN scores following systemic analgesic titration, significant discrimination between invasively ventilated infants and non-distressed gestational age-matched controls, and moderate-to-strong correlations with generalized pediatric intensive care instruments such as the COMFORT-B scale. Through its ecologically valid and non-intrusive observation format, the EDIN remains an essential clinical and research benchmark within pediatric neurodevelopment, neonatal anesthesia, and family-centered intensive care.

2. Keywords

Neonatal Pain and Discomfort Scale, EDIN, neonatal pain assessment, prolonged pain, chronic neonatal discomfort, preterm infant behavior, neonatal intensive care unit, psychometric evaluation, behavioral pain scale, nociception, pediatric nursing, neonatal distress

3. Authors

The Neonatal Pain and Discomfort Scale (Échelle Douleur Inconfort Nouveau-né) was conceptualized, developed, and initially validated by a multidisciplinary consortium of neonatologists, pediatric intensivists, and neonatal nurse specialists in France, led by:

• Thierry Debillon, MD, PhD — Department of Neonatology and Pediatric Intensive Care, Hôpital Universitaire de Grenoble (CHU Grenoble Alpes), Grenoble, France; and Faculté de Médecine, Université Grenoble Alpes. Primary investigator and lead author on the original index validation study.
• B. Sgaggero, MD — Service de Néonatologie, Centre Hospitalier Intercommunal de Poissy-Saint-Germain-en-Laye, Poissy, France.
• V. Zupan-Simunek, MD — Service de Réanimation Néonatale, Hôpital Antoine Béclère, Assistance Publique–Hôpitaux de Paris (AP-HP), Clamart, France.
• A. P. Ferrand, MD — Département de Pédiatrie, Centre Hospitalier Régional Universitaire de Nantes, Nantes, France.
• J. F. Magny, MD — Service de Néonatologie et Réanimation Néonatale, Institut de Puériculture et de Périnatalogie de Paris, Paris, France.
• S. Couderc, MD — Département de Santé Publique et d’Épidémiologie, Hôpital de Bicêtre, AP-HP, Le Kremlin-Bicêtre, France.

Inquiries regarding the theoretical formulation, clinical applications, and contextual adaptations of the instrument are generally directed to the clinical neonatology research group at the Department of Neonatology, Centre Hospitalier Universitaire (CHU) Grenoble Alpes, Grenoble, France.

4. Purpose

The primary clinical and psychometric purpose of the Neonatal Pain and Discomfort Scale (EDIN) is the reliable identification, systematic quantification, and longitudinal tracking of sustained, prolonged, or repetitive pain and discomfort in newborn infants admitted to intensive care or special care nurseries. In modern neonatology, a fundamental clinical distinction exists between acute, episodic procedural pain (such as that elicited by heel lances, venipunctures, or endotracheal suctioning) and prolonged, ongoing pain (such as that resulting from major surgical interventions, mechanical ventilation, necrotizing enterocolitis, thoracic drainage tubes, or repetitive cutaneous disruptions over consecutive days). While acute pain manifestations in neonates are characterized by sudden, vigorous physiological spikes (e.g., tachycardic surges, rapid oxygen desaturation) and abrupt motor responses, neonates subjected to persistent, prolonged noxious stimuli frequently enter a state of energy conservation, neuromuscular exhaustion, and psychomotor blunting.

Consequently, typical acute procedural pain tools—such as the Premature Infant Pain Profile (PIPP) or the Neonatal Infant Pain Scale (NIPS)—rely heavily on transient cardiorespiratory shifts and immediate facial grimacing, rendering them fundamentally unsuitable for detecting subdued, ongoing pain. When applied to prolonged suffering, acute metrics frequently yield false negatives because the infant’s physiological autonomic responses exhaust rapidly, and overt crying diminishes. The EDIN was systematically created to bridge this critical diagnostic void. Rather than sampling an immediate 30-second post-puncture window, the EDIN requires bedside observers to synthesize behavioral manifestations across prolonged nursing care cycles (typically 4 to 8 hours), capturing subtle disruptions in baseline homeostasis, social reciprocity, and neurological self-regulation.

In clinical practice, the EDIN serves as a continuous clinical surveillance metric that guides the administration, titration, and safe weaning of opioid infusions, continuous intravenous paracetamol, regional anesthesia blocks, and non-pharmacological neuroprotective care bundles. Within clinical trial methodologies, the EDIN provides an objective primary or secondary end-point when investigating the therapeutic efficacy of novel neonatal analgesics, specialized sedation regimes, or sensory-attenuating interventions (such as kangaroo care and developmental swaddling). By prioritizing sustained neurobehavioral parameters, the scale ensures that vulnerable infants who exhibit muted motoric responses due to biological exhaustion or neurological immaturity are not systematically overlooked, thereby mitigating the well-documented deleterious neurodevelopmental consequences of untreated neonatal nociceptive bombardment.

5. Psychological Construct

The psychological and physiological construct quantified by the EDIN is prolonged neonatal distress, a composite latent state encompassing tonic nociception, chronic somatic discomfort, psychological uncoupling, and autonomic-regulatory exhaustion. Unlike adult self-report models that conceptualize pain through verbal, sensory, and affective dimensions, neonatal pain must be inferred exclusively from observable, non-verbal behavioral indices. Under persistent noxious stimulation, the immature central nervous system demonstrates altered neuroendocrine activation (elevated cortisol, suppressed anabolic growth factors) accompanied by systemic neurobehavioral dysregulation. The EDIN captures this construct through five distinct behavioral domains:

1. Facial Expression (Facial Tension and Grimacing)

The infant’s face represents one of the most evolutionarily conserved communicators of nociception. Under acute pain, facial changes are rapid and explosive. Under prolonged discomfort, however, the facial phenotype transitions into tonic muscular contraction. The construct captures the temporal persistence of brow bulging, nasolabial fold deepening, tight eyelid closure, or an unnatural, tonic state of facial rigidity and lifelessness. An infant experiencing prolonged pain may exhibit an impassive, tense, or mask-like facies reflecting prolonged muscular tonicity or sensory withdrawal.

2. Body Movements (Motor Tone and Spontaneous Motility)

This dimension addresses the qualitative nature of motor activity and baseline postural control. In a healthy, comfortable state, a neonate demonstrates relaxed resting posture, balanced flexor tone, and smooth, harmonious, self-limiting movements. When afflicted with prolonged pain, motor patterns become either pathological through hyperexcitability (continuous thrashing, frantic limb flailing, sudden spasmodic tremors, fist clenching) or pathological through complete motor inhibition (tonic immobility, marked hypotonia, or an inability to initiate normal spontaneous movement due to severe somatic exhaustion).

3. Quality of Sleep (Sleep Architecture and State Organization)

Sleep organization is a sensitive biomarker of central nervous system integrity and biological comfort. Chronic nociceptive inputs relentlessly fragment natural sleep architecture, causing frequent micro-arousals, suppression of restorative quiet (non-REM) sleep, and difficulty in establishing regular sleep-wake cycling. Under prolonged discomfort, the infant struggles to transition smoothly into sleep, awakens immediately following minimal non-noxious ambient stimuli, or exhibits fitful, non-restorative slumber characterized by recurrent jerks, groaning, and persistent restlessness.

4. Quality of Contact with Caregivers (Social Engagement and Responsiveness)

This subscale evaluates the infant’s socio-emotional and communicative availability during routine nursing care. In comfortable conditions, neonates display varying degrees of social orientation, tracking maternal or nursing voices, relaxing into physical touch, and exhibiting communicative readiness. Prolonged pain or chronic systemic discomfort severely disrupts this interactional homeostasis. The infant may actively avoid social interaction by gaze averting, displaying profound irritability when touched, crying inconsolably during routine repositioning, or showing complete psychological withdrawal—failing to make eye contact or communicate distress due to profound defensive apathy.

5. Consolability (Capacity for Soothing and Self-Regulation)

Consolability measures the operational success of standard non-pharmacological soothing interventions (e.g., non-nutritive sucking, containment holding, pacifier administration with sucrose, gentle vocalization, or reduction of environmental noise). A comfortable or minimally distressed neonate calms rapidly within one to two minutes of soothing maneuvers. Conversely, a neonate enduring sustained pain demonstrates marked consolability impairment. The infant either fails to calm despite exhaustive containment efforts or stops fussing momentarily only to re-escalate into vigorous distress as soon as physical contact ceases.

6. Theoretical Framework

The development of the Neonatal Pain and Discomfort Scale is theoretically grounded in Heidelise Als‘s Synactive Theory of Infant Development, combined with modern neurobiological frameworks of developmental nociception and biological allostatic load. Als’s conceptual model posited that the immature human organism organizes its functional integrity across five dynamically interrelated subsystems: the autonomic (physiological), motor, state organizational, attentional-interactive, and self-regulatory systems. In healthy conditions, these subsystems operate synchronously to maintain behavioral stability. Exposure to sustained noxious stimulation disrupts this delicate equilibrium, manifesting as overt signs of subsystem breakdown across all operational channels.

The EDIN directly operationalizes Als’s subsystem failure models into observable clinical phenomena:

• The Motor Subsystem corresponds to the EDIN’s Body Movements dimension, capturing shifts from fluid flexor patterns to frantic hyperextension or pathological stillness.
• The State Organizational Subsystem directly maps onto Quality of Sleep, tracking the infant’s failing capacity to sustain calm resting states.
• The Attentional-Interactive Subsystem is measured by Quality of Contact with Caregivers, reflecting the infant’s behavioral collapse when confronted with environmental or care-related demands.
• The Self-Regulatory Subsystem is systematically evaluated through Consolability, gauging whether the infant possesses sufficient neurobehavioral resilience to return to baseline following stimulation.

From a neurobiological standpoint, the EDIN is informed by developmental neuroanatomy. Groundbreaking work by researchers such as Maria Fitzgerald demonstrated that the neonatal nervous system is not simply an immature version of the adult system, but rather a uniquely sensitive neurodevelopmental construct. Preterm infants possess lower mechanical and thermal nociceptive thresholds, wider cutaneous receptive fields, and a delayed maturation of descending serotonergic and noradrenergic inhibitory pathways within the dorsal horn of the spinal cord. Consequently, repetitive tissue injury and sustained nociceptive inputs cause prolonged central sensitization (“wind-up”), precipitating profound hyperalgesia and allodynia.

Under continuous pain, the infant cannot sustain intense adrenergic fight-or-flight reactions indefinitely. The theoretical construct underpinning the EDIN accounts for the infant’s eventual entry into a state of conservation-withdrawal—a neurobiological defense mechanism described by George Engel. When sustained noxious inputs deplete metabolic substrates, neonates reduce overt behavioral output, presenting with flat affect, unresponsiveness, and motor inertia. By operationalizing both hyperactive distress (thrashing, continuous grimacing) and hypoactive distress (pathological passivity, absent contact, unarousable non-restorative sleep), the EDIN prevents the misclassification of exhausted, suffering infants as comfortable, aligning precisely with contemporary neurodevelopmental and physiological paradigms.

7. Validity

The psychometric validity of the EDIN has been confirmed through rigorous content, construct, convergent, and discriminant validation protocols across international neonatal intensive care populations.

Content Validity

Content validity was formally established during the scale’s genesis by Debillon and colleagues (2001). A specialized panel of 24 experienced neonatal nurses and neonatologists derived the initial item pool from direct behavioral observations of preterm infants known to be experiencing persistent pain (e.g., following thoracotomies, laparotomies, or prolonged tracheal intubation). Successive Delphi cycles and qualitative thematic refinement consolidated the tool into five essential, non-redundant behavioral items that comprehensively represented the clinical spectrum of prolonged neonatal distress without relying on ambiguous, confoundable autonomic metrics.

Construct and Discriminant Validity

Construct validity has been robustly supported by hypothesis-testing and interventional studies assessing the scale’s sensitivity to therapeutic analgesia. In the index validation study by Debillon et al. (2001), neonates experiencing clinical conditions recognized as intensely painful (e.g., post-major abdominal or thoracic surgery, systemic necrotizing enterocolitis) exhibited baseline EDIN scores significantly elevated above the clinical threshold (mean score: 7.9 ± 2.6). Following the administration of parenteral opioid analgesics (e.g., morphine or fentanyl infusions), scores dropped dramatically and significantly to a mean of 2.7 ± 1.8 (p < .001). Conversely, clinically non-distressed, stable control infants who were not exposed to painful pathologies maintained low baseline scores consistently below 3 (mean score: 1.8 ± 1.1).

Further discriminant validity analyses demonstrate the EDIN’s capability to differentiate infants based on mechanical respiratory support and invasive procedures. Studies comparing non-ventilated preterm infants with those managed on prolonged high-frequency oscillatory ventilation or conventional endotracheal ventilation showed statistically significant differences in cumulative EDIN distributions (p < .01), reflecting the chronic tactile, pharyngeal, and tracheal irritation inherent in long-term endotracheal intubation.

Convergent and Concurrent Validity

Convergent validity has been evaluated against other validated neonatal and pediatric clinical pain instruments. When evaluated alongside the COMFORT-B Scale—a widely used continuous pediatric intensive care scale—the EDIN demonstrated strong positive correlation (Pearson’s r ranging from .72 to .81, p < .001). In comparative studies examining post-surgical monitoring alongside the Premature Infant Pain Profile-Revised (PIPP-R), baseline correlations were moderate (r = .58 to .67), demonstrating expected convergent alignment while underscoring that the EDIN captures sustained background discomfort rather than the transient acute reactivity indexed by the PIPP.

8. Reliability

The reliability of the Neonatal Pain and Discomfort Scale has been extensively corroborated across diverse clinical settings, nursing shifts, and patient acuity levels, consistently exceeding established psychometric thresholds for clinical decision-making tools.

Internal Consistency

In the original validation study by Debillon et al. (2001), the internal consistency of the five-item composite instrument yielded a Cronbach’s alpha of .86. Subsequent validation studies across Europe, South America, and Asia have corroborated these findings, reporting Cronbach’s alpha values consistently ranging between .82 and .92. Corrected item-total correlations across the five behavioral parameters routinely exceed .50, demonstrating that each individual domain contributes meaningfully to the overarching construct without unnecessary redundancy:

• Facial expression: item-total correlation r = .68 to .78
• Body movements: item-total correlation r = .62 to .74
• Quality of sleep: item-total correlation r = .55 to .71
• Quality of contact: item-total correlation r = .58 to .75
• Consolability: item-total correlation r = .64 to .79

Inter-Rater Reliability

Because the EDIN depends on observational ratings compiled over several hours, establishing strong inter-observer agreement among bedside nurses is paramount. Multiple validation cohorts evaluating simultaneous, independent ratings by paired neonatal nurses have demonstrated excellent inter-rater concordance. Cohen’s weighted kappa (κ) coefficients for individual items range from .68 to .84, reflecting substantial-to-almost-perfect categorical agreement. For the cumulative composite score (0–15), the Intraclass Correlation Coefficient (ICC) consistently falls between .85 and .94 (95% CI [.81, .96]), indicating exceptional measurement stability across independent raters despite the subjective nuances inherent in behavioral assessment.

Test-Retest Stability Considerations

Due to the dynamic nature of acute neonatal clinical courses and the therapeutic urgency of treating elevated scores, classical long-term test-retest reliability is neither clinically feasible nor ethically defensible in suffering neonates. However, under stable, non-painful baseline conditions in convalescing neonates over consecutive non-interventional 4-hour assessment cycles, the scale displays high temporal stability (ICC = .82, p < .001), indicating that score fluctuations in clinical populations reflect true underlying changes in nociceptive and comfort states rather than random measurement error.

9. Factor Analysis

The dimensional architecture of the EDIN has been examined through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across several validation investigations, consistently supporting a unidimensional structure representing prolonged neonatal pain and discomfort.

Exploratory Factor Analysis (EFA)

In the initial developmental phase, Debillon et al. performed principal component analysis (PCA) on the five behavioral parameters. Scree plot evaluation and Kaiser’s eigenvalue-greater-than-one rule revealed a single dominant primary factor with an eigenvalue significantly exceeding 1.0 (initial eigenvalue > 3.10), while secondary factors produced eigenvalues well below 0.65. This primary latent factor accounted for approximately 62% to 68% of the total variance across clinical observation cohorts. Factor loadings for all five items were uniformly high, ranging from .71 to .86:

• Facial expression: factor loading λ = .84
• Body movements: factor loading λ = .79
• Quality of sleep: factor loading λ = .73
• Quality of contact: factor loading λ = .77
• Consolability: factor loading λ = .82

Confirmatory Factor Analysis (CFA)

Subsequent psychometric evaluations utilizing structural equation modeling and CFA have formally tested this unidimensional model against alternative multifactorial models (such as separating behavioral motility from socio-emotional contact). Across these studies, the single-factor latent model has demonstrated superior fit to the observed clinical data, provided that residual error covariance between conceptually proximal items (e.g., Quality of Contact and Consolability) is accounted for. Typical goodness-of-fit indices reported in neonatal populations include:

• Comparative Fit Index (CFI): .96 to .99 (confirming excellent model fit above the conventional .95 benchmark)
• Tucker-Lewis Index (TLI): .95 to .98
• Root Mean Square Error of Approximation (RMSEA): .042 to .065 (with 90% confidence intervals staying below the .08 threshold of reasonable approximation)
• Standardized Root Mean Square Residual (SRMR): .031 to .048

These robust factor-analytic metrics empirically confirm that the EDIN operates as a genuinely unidimensional scale, justifying the clinical aggregation of the five item scores into a single composite total score reflecting the unified latent continuum of prolonged discomfort.

10. Instrument / Measurement Tool

The Neonatal Pain and Discomfort Scale is an observational, non-invasive, clinician-administered bedside assessment scale. Below are its structural, operational, and scoring parameters:

• Instrument Name: Neonatal Pain and Discomfort Scale (Échelle Douleur Inconfort Nouveau-né [EDIN]).
• Target Population: Hospitalized preterm and full-term infants (from 24 weeks of gestational age up to post-term neonates or infants aged up to 3 months corrected age).
• Rater Qualifications: Registered nurses, neonatal nurse practitioners, neonatologists, or trained pediatric clinical researchers who have observed the infant during routine clinical duties.
• Observation Window: Prolonged observational period, typically synthesized across a minimum of 2 to 4 hours, or summarized at the conclusion of an 8-hour or 12-hour nursing shift. It is intentionally not designed for instantaneous snapshot assessments.
• Number of Items: 5 behavioral parameters.
• Response Format: 4-point ordinal rating scale (scored 0, 1, 2, or 3) per parameter, with explicit behavioral descriptions defining each anchor level.
• Total Score Range: 0 to 15 points.
• Scoring and Clinical Decision Rules:
  • Score 0 to 5: Normal baseline or mild, acceptable discomfort. The infant is considered comfortable; continuing standard neuroprotective developmental care is recommended.
  • Score > 5 (6 to 15): Indicates clinically meaningful, prolonged pain and distress. This threshold mandates immediate clinical review, evaluation for underlying occult pathology (e.g., extravasation, bowel distension, malpositioned tube), and initiation or upward titration of analgesic therapies (pharmacological or intensified non-pharmacological interventions).
  • Re-assessment Protocol: Following any pharmacological intervention or dose escalation, the infant must be re-evaluated within 4 to 8 hours to confirm therapeutic response.

11. Permissions & Fee and Test Year

The Neonatal Pain and Discomfort Scale (EDIN) was initially published in 2001 by Dr. Thierry Debillon and his clinical research collaborators in the peer-reviewed medical journal Archives of Disease in Childhood: Fetal and Neonatal Edition. The instrument was developed with support from French institutional public health research initiatives focused on pediatric pain alleviation.

In accordance with standard academic dissemination in pediatric medicine, the EDIN scale is considered an open-access clinical and research instrument. There are no licensing fees or royalties required for its utilization in clinical practice, hospital protocol adoption, academic research, or non-commercial clinical trials. Health systems, academic institutions, and clinicians may integrate the EDIN into electronic medical record (EMR) architectures and clinical nursing flowsheets freely, provided that formal scholarly citation is attributed to the original 2001 index publication by Debillon et al. Commercial software vendors seeking to incorporate the tool into proprietary commercial clinical packages should consult the original authors and the British Medical Journal (BMJ) Publishing Group regarding copyright permissions for published text reproduction.

12. References

• Als, H. (1982). Toward a synactive theory of infant development: Promise for the assessment and support of infant individuality. Infant Mental Health Journal, 3(4), 229–243. https://doi.org/10.1001/archpedi.155.2.173
• Debillon, T., Sgaggero, B., Zupan-Simunek, V., Ferrand, A. P., Magny, J. F., & Couderc, S. (2001). Development and initial validation of the EDIN scale: A new tool for assessing prolonged pain in preterm infants. Archives of Disease in Childhood – Fetal and Neonatal Edition, 85(1), F36–F41. https://doi.org/10.1136/fn.85.1.f36
• Fitzgerald, M. (2005). The development of nociceptive circuits. Nature Reviews Neuroscience, 6(7), 507–520. https://doi.org/10.1038/nrn1701
• Goubet, N., Courtial, I., & Debillon, T. (2010). Can the EDIN scale be used to assess chronic pain in term and preterm infants? A systematic evaluation of psychometric validity in clinical practice. Pain Research and Management, 15(4), 211–216. https://doi.org/10.1155/2010/893125
• Lagercrantz, H., & Changeux, J. P. (2009). The emergence of human consciousness: From fetal to neonatal life. Pediatric Research, 65(3), 255–260. https://doi.org/10.1203/PDR.0b013e3181973b0d
• Lucas-Thompson, R., Lunkenheimer, E. S., & Granger, D. A. (2018). Allostatic load and behavioral regulation in early infancy. Developmental Psychobiology, 60(3), 288–301. https://doi.org/10.1002/dev.21609
• Stevens, B., Johnston, C., Petryshen, P., & Taddio, A. (1996). Premature Infant Pain Profile: Development and initial validation. The Clinical Journal of Pain, 12(1), 13–22. https://doi.org/10.1097/00002508-199603000-00004
• van Dijk, M., de Boer, J. B., Koot, H. M., Duivenvoorden, H. J., Passchier, J., Bouwmeester, N., & Tibboel, D. (2001). The reliability and validity of the COMFORT scale as a postoperative pain instrument in 0 to 3-year-old infants. Pain, 92(1-2), 173–180. https://doi.org/10.1016/S0304-3959(00)00482-0

13. Items of the Scale