Journal Logo


Using only behaviours to assess infant pain: a painful compromise?

Pillai Riddell, Rebeccaa; Fitzgerald, Mariab; Slater, Rebeccahc; Stevens, Bonnied; Johnston, Celestee; Campbell-Yeo, Marshaf

Author Information
doi: 10.1097/j.pain.0000000000000598
  • Free

In this issue of PAIN®, Välitalo and colleagues' report on an Item Response Theory analysis examining the utility of behavioural and physiological measures of the autonomic nervous system.18 Based on their findings, they discourage the use of physiological indicators when assessing the pain of mechanically ventilated infants and posit that infant pain assessment could primarily rely on infant behaviours. Interestingly, such physiological indicators are often used within popular clinical assessment measures. Thus, their conclusions represent a significant departure from current practice. If clinicians go on to use unvalidated modifications of existing assessment measures, it could be damaging to infants in care. However, their article draws our attention to unresolved challenges within infant pain assessment.

Pain is a sensory and emotional phenomenon8 that is ubiquitous; yet the understanding of pain in another person is elusive.8 Despite common features of basic nociceptive processing, pain is subjective. The transduction of a noxious stimulus into pain experience is dependent on individual physiology, personal experiences, and social context. The challenge of infant pain assessment is further compounded because validation cannot be obtained from the preverbal infant. In addition, the steep trajectory of cognitive and biological development likely means that the experience of pain changes over the first years of life.4,15

Questioning the utility of physiological measures is not new.11,13 Physiological measures always will lack specificity to pain because physiological responses do not only change as a direct response to noxious stimulation. Infant physiological responses to noxious stimuli also have not been properly examined longitudinally to provide valid norms for comparison.20 Yet, despite these challenges, physiological indicators are the cornerstones of most pain assessment tools in hospitalized infants. Who is best positioned to assess infant pain, and how should this assessment be done?

All things being equal, parents are the caregivers who have greatest exposure to and insight into their infant's behavior.9 Interpreting the needs of an infant based on behavior is fundamental to the provision of infant care. For parents, the most readily available method of assessing their infant's pain is to rely on behaviours such as facial expression, body movements, and crying. Parents report that these indicators are among the most important and salient ones in making pain judgments.12 Interestingly, however, both quasiexperimental studies and multivariate modeling studies have shown that the preponderance of variance in parental pain judgments is not based on infant behaviours.12–14

Similar concerns also arise as to the actual basis of nurses and physicians judgments of pain.14 One cannot use visual analogue scale ratings provided by health care providers as a gold standard proxy to evaluate the validity of an infant pain measure without recognizing how intrinsic factors bias their ratings, such as age, sex, culture, profession.13

To reduce bias, clinicians, and clinician scientists are increasingly emphasizing the need for more specialized training and formal assessment tools for assessing pain in young children. Multidimensional or composite pain tools (such as the PIPP-R5) integrate physiological and behavioural measures. Indeed, the objectivity of physiological measures is alluring. In a basic sense, the measurement of the actual physiological processes is highly reliable and valid.

It is clear that the sole dependence on behavioural indicators (such as calmness or facial agitation, as suggested by Välitalo and colleagues) also has limitations. The specificity of behavioural pain measures has been questioned,2 as has been the underlying cognitive ability of the young infant to discern and express pain differently from other negative affect states.1 Of further note, one behavioural omission was the preterm hand behaviours that have been studied in detail by Holsti and Grunau through their work on the assessment tool, Behavioural Indicators of Infant Pain.7 In addition, relationships between nociceptive central nervous system activity and different types of infant behaviour are variable, and this has cast doubt on the use of behaviour as a surrogate measure of infant pain.6,17

The challenge of infant pain assessment was the recent subject of an international consensus meeting that brought together basic scientists, basic behavioural scientists, clinician scientists, and clinicians (Determining a Comprehensive Approach to Measuring Pain in Neonates and Infants, November 2014). At this meeting, there was a clear consensus that cortical, physiological, and behavioural measures of pain do not converge to an extent that would exude confidence. From a purely epistemological vantage point, it seems that basic and clinical scientists may not be measuring the same phenomenon, and simplistic attempts to converge their measurement approaches may not be useful.

Basic scientists are trying to establish the fundamental pathways by which noxious events are transmitted and processed by the infant central nervous system, whereas behavioural scientists address how this process plays out in the behavior and social interactions of the infant. Both approaches build on an underlying framework of academic neuroscience and psychology. The clinician who works with infants, however, is necessarily focused on the problem of inadequately treated pain and seeks a safe and practical solution. Which group (basic scientists or clinicians) is really studying “infant pain?” The answer is both of them. Although, some would argue that, given the hallmark of pain being a subjective experience, the answer could be none of them because we do not have access to their verbal report.

The pragmatic and urgent need for infant pain assessment does not allow the luxury of prolonged theoretical debate.10,16 Clinicians must still take action in the absence of a feasible, valid, and reliable measure of infant pain that adequately takes into account not only steep development within infancy but also the sensory, social, and affective dimensions of pain. Despite the lack of specificity to pain-related distress and the lack of convergence with changes in pain-related brain activity, clinicians would not be completely remiss to follow the suggestion of Välitalo et al. to use behavioural indicators of pain. However, they should be able to make nuanced pain treatment decisions in the context of very fundamental limitations to this approach.

For basic and behavioural scientists, the story goes on as we are actually at the very beginning of understanding the complexity of pain within the developing nervous system.3,4 The development of a clinically useful pain assessment tool may depend on our understanding of how nociceptive stimulation alters activity across all levels of the peripheral and central nervous system and how this activity is linked to overt behaviours seen in the clinical setting.

Conflict of interest statement

The authors have no conflicts of interest to declare.


The “Determining a Comprehensive Approach to Measuring Pain in Neonates and Infants: a Consensus Meeting” (Canadian Institutes of Health Research Planning Grant, # 218078) participants were: Simon Beggs, Britney Benoit, Janet Brown, Marsha Campbell-Yeo (Co-I), Ricardo Carbajal, Eva Cignacco, Ken Craig, Maria Fitzgerald, Sharyn Gibbins, Ruth Grunau, Denise Harrison, Liisa Holsti, Celeste Johnston (Co-I), Beatriz Linhares, Carol McNair, Steven Miller, Rebecca Pillai Riddell (Co-I), Manon Ranger, Vibhutti Shah, Rebeccah Slater, Bonnie Stevens (PI), Anna Taddio, Monique van Dijk, and Janet Yamada (Co-I).


[1]. Ahola Kohut S, Pillai Riddell P, Flora DB, Oster H. A longitudinal analysis of the development of infant facial expressions in response to acute pain: immediate and regulatory expressions. PAIN 2012;153:2458–65.
[2]. Ahola Kohut S, Pillai Riddell R. Does the Neonatal Facial Coding System differentiate between infants experiencing pain-related and non-pain-related distress? J Pain 2009;10:214.
[3]. Fitzgerald M. Central nociceptive pathways and descending modulation. In McGrath PJ, Stevens BJ, Walker SM, Zempsky WT, editors. Oxford Textbook of Paediatric Pain. Oxford University Press, 2013.
[4]. Fitzgerald M. What do we really know about newborn infant pain? Exp Physiol 2015;100:1451–7.
[5]. Gibbins S, Stevens BJ, Yamada J, Dionne K, Campbell-Yeo M, Lee G, Caddell K, Johnston C, Taddio A. Validation of the Premature Infant Pain Profile-Revised (PIPP-R). Early Hum Dev 2014;90:189–93.
[6]. Hartley C, Goksan S, Poorun R, Brotherhood K, Mellado GS, Moultrie F, Rogers R, Adams E, Slater R. The relationship between nociceptive brain activity, spinal reflex withdrawal and behaviour in newborn infants. Sci Rep 2015;5. doi: 10.1038/srep12519.
[7]. Holsti L, Grunau RE. Initial validation of the behavioral indicators of infant pain (BIIP). PAIN 2007;132:264–72.
[8]. International association for the study of pain: IASP taxonomy. Available at: Accessed December 9, 2015.
[9]. Jiang S, Warre R, Xianming Q, O'Brien K, Lee SK. Parents as practitioners in preterm care. Early Hum Dev 2014;90:781–5.
[10]. Johnston C, Barrington KJ, Taddio A, Carbajal R, Filion F. Pain in Canadian NICUs: have we improved over the past 12 years? Clin J Pain 2011;27:225–32.
[11]. Johnston CC, Stremler RL, Stevens BJ, Horton LJ. Effectiveness of oral sucrose and simulated rocking on pain response in preterm neonates. PAIN 1997;72:193–9.
[12]. Pillai Riddell R, Badali MA, Craig KD. Parental judgments of infant pain: importance of perceived cognitive abilities, behavioral cues and contextual cues. Pain Res Manag 2004;9:73–80.
[13]. Pillai Riddell R, Craig KD. Judgments of infant pain: the impact of caregiver identity and infant age. J Pediatr Psychol 2007;32:501–11.
[14]. Pillai Riddell R, Flora DB, Stevens S, Greenberg S, Garfield H. The role of infant pain behaviour in predicting parent pain ratings. Pain Res Manag 19;2014:e124–32.
[15]. Pillai Riddell R, Flora DB, Stevens SA, Stevens B, Cohen LL, Greenberg S, Garfield H. Variability in infant acute pain responding meaningfully obscured by averaging pain responses. PAIN 2013;154:714–21.
[16]. Slater R, Cantarella A, Franck L, Meek J, Fitzgerald M. How well do clinical pain assessment tools reflect pain in infants? PLoS Med 5;2008:e129.
[17]. Stevens BJ, Abbott LK, Yamada J, Harrison D, Stinson J, Taddio A, Barwick M, Latimer M, Scott SD, Rashotte J, Campbell F. Epidemiology and management of painful procedures in children in Canadian hospitals. Can Med Assoc J 2011;183:E403–10.
[18]. Välitalo PAJ, van Dijk M, Krekels EHJ, Gibbins S, Simons SH, Tibboel D, Knibbe CAJ. Pain and distress caused by endotracheal suctioning in neonates is better quantified by behavioural than physiological items: a comparison based on Item Response Theory modelling. PAIN 2016;157:1611–7.
[19]. Verriotis M, Fabrizi L, Lee A, Ledwidge S, Meek J, Fitzgerald M. Cortical activity evoked by inoculation needle prick in infants up to one-year old. PAIN 2015;156:222–30.
[20]. Waxman J, Pillai Riddell RR, Pinhasov A, Schmidt LA, Tablon P. development of cardiovascular indices of acute pain responding in infants: a systematic review and meta-analysis. Pain Res Manag 2016,
© 2016 International Association for the Study of Pain