609
ACAD EMERG MED • June 2002, Vol. 9, No. 6 • www.aemj.org
Parents and Practitioners Are Poor Judges of Young Children’s Pain Severity Adam J. Singer, MD, Janet Gulla, RN, Henry C. Thode Jr., PhD Abstract Objective: Visual analog pain scales are reliable measures in older children and adults; however, pain studies that include young children often rely on parental or practitioner assessments for measuring pain severity. The authors correlated patient, parental, and practitioner pain assessments for young children with acute pain. Methods: This was a prospective, descriptive study of a convenience sample of 63 emergency department patients aged 4–7 years, with acute pain resulting from acute illness or painful invasive procedures. A trained research assistant administered a structured pain survey containing demographic and historical features to all parents/guardians. Children assessed their pain severity using a validated ordinal scale that uses five different faces with varying degrees of frowning (severe pain) or smiling (no pain). Each face was converted to a numeric value from 0 (no pain) to 4 (severe pain). Parents and practitioners independently assessed their child’s pain using a validated 100-mm visual analog scale (VAS) marked ‘‘most pain’’ at the high end. Pairwise correlations between child, parent, and practitioner pain assess-
ments were performed using Spearman’s or Pearson’s test as appropriate. The association between categorical data was assessed using 2 tests. Results: Sixty-three children ranging in age from 4 to 7 were included. Mean age (⫾SD) was 5.7 (⫾1.1); 42% were female. Fifty-seven successfully completed the face scale. The distribution of the children’s scores was 0–17%, 1–9%, 2–30%, 3–14%, and 4–30%. Mean parental and practitioner scores (⫾SD) on the VAS were 61 (⫾26) mm and 37 (⫾26) mm, respectively (maximal = 100 mm). Correlation between child and parent scores was 0.47 (p < 0.001). Correlation between child and practitioner scores was 0.08 (p = 0.54). Correlation between parent and practitioner scores was 0.04 (p = 0.001). Conclusions: There is poor agreement between pain ratings by children, parents, and practitioners. It is unclear which assessment best approximates the true degree of pain the child is experiencing. Key words: pain scales; pediatric; correlation; Smiley Analog Scale; visual analog scale; judgment. ACADEMIC EMERGENCY MEDICINE 2002; 9:609–612.
It has been suggested that research on pain control in children has traditionally been limited by the erroneous notion that children experience less pain than adults due to neurological immaturity.1 Research has also been limited by the difficulty of assessing the pain experience in children. This situation was improved in the 1980s by data indicating that infants and young children clearly experience pain.2,3 Also, development of age-appropriate scales has allowed better assessment of pain in children.4,5 Although many health care practitioners and parents believe that they can reliably judge the severity of pediatric pain, to the best of our knowledge, this assumption has not been validated. The objective of the current study was to determine whether assessments of pain severity by children aged 4–7 years correlate with similar assessments made by their parents and health care practitioners.
METHODS
From the Department of Emergency Medicine, State University of New York, Stony Brook, NY. Received November 2, 2001; revision received December 21, 2001; accepted January 10, 2002. Presented in part at the SAEM annual meeting, Atlanta, GA, May 2001. Address for correspondence and reprints: Adam J. Singer, MD, Department of Emergency Medicine, University Hospital and Medical Center, L4-515, Stony Brook, NY 11794-7400. E-mail:
[email protected].
Study Design. A survey was conducted to correlate pain severity assessments of children with those of their parents and health care practitioners. The study was approved by both the Institutional Review Board and the Committee on Research Involving Human Subjects prior to initiation. Verbal consent was obtained from all patients’ guardians. Study Setting and Population. The study was conducted in the Emergency Department (ED) of the State University of New York at Stony Brook, a tertiary care university-based medical center with an annual census of 65,000. English-speaking children between the ages of 4 and 7 years with an acute painful condition with a duration of less than 24 hours, or those undergoing a painful procedure in the ED, were eligible for enrollment. The study population comprised a convenience sample of eligible patients who presented to the ED when a trained research assistant was present.6 Research assistants are present in the ED seven days a week between 8:00 AM and midnight. Study Protocol. Patient demographic and clinical information was collected using a standardized
610 data collection form. Specific demographic and clinical information collected included the patient’s age, gender, chief complaint, and nature of any diagnostic or therapeutic procedures performed. Demographic information concerning the enrolled childrens’ parents and practitioners was also collected. Children assessed their pain severity using a validated ordinal scale7 [the Smiley Analog Scale (SAS)] that uses five different faces with varying degrees of frowning (severe pain) or smiling (no pain, Fig. 1). The child was asked to point to the face that best described how he or she was feeling at that moment. Each face was converted to a numeric value from 0 (no pain) to 4 (severe pain). The same faces scale has also been used in other studies.8 Similar facial expression scales have also been used and shown to have construct validity.1 Parents and practitioners independently assessed their child’s pain using a validated 100-mm visual analog scale (VAS) marked ‘‘most pain’’ at the high end. Parents and practitioners were specifically masked to their children’s assessments. When using the VAS, patients were asked to mark the line at a position between the two extremes that represented the level of their child’s/patient’s pain. The pain score was obtained by measuring the distance in mm from the extreme end marked ‘‘no pain.’’ For painful procedures, pain assessments were performed immediately after completion of the procedure. For acute painful conditions, pain assessments were obtained immediately after the practitioner had evaluated the child. Data Analysis. All data were entered into SPSS 8.5 for Windows (SPSS, Inc., Chicago, IL) for statistical analysis. Categorical data are presented as percent frequency of occurrence and continuous data are presented as means and standard deviations. Pair-
Singer et al. • JUDGMENT OF CHILDREN’S PAIN SEVERITY
wise correlations between child, parent, and practitioner pain assessments were performed using Spearman’s or Pearson’s test for ordinal and continuous data, respectively. The association between categorical data was assessed using the 2 test. Assuming a correlation near 0.80, a sample size of 60 would provide accuracy of the estimate to have a correlation coefficient confidence interval of ⫾0.10.9 A correlation of 0.80 is considered excellent,10 and 23 mm was the standard deviation for pain scores after intravenous catheterization.11
RESULTS During the study period, 63 children ranging in age from 4 to 7 years were included. Most, but not all, of the children successfully completed the SAS (57/ 63). Their mean age was 5.7 (⫾1.1), 42% were female, and 80% were white. The mean age of the children who did not complete the SAS was not significantly different from that of the children who successfully completed the SAS (5.3 vs. 5.7, Wilcoxon test p = 0.42). The children who did not complete the SAS are excluded from all further analyses. Of 60 children who completed the SAS, three were excluded due to incorrect use of the forms by the investigator. Of the 57 children, 13 underwent a painful procedure (intravenous catheterization, 5; suturing, 5; phlebotomy, 1; intramuscular injection, 1; suture removal, 1) and 44 had a painful medical condition. In these children the pain was generalized (2) or located on the extremities (18), head and neck (17), or torso (7). Parental pain assessments were performed by 48 mothers, 5 fathers, and 4 guardians. Their mean age was 36 (⫾7); 86% were female. Health care practitioner pain assessments were performed by 25 physicians, 23 nurses, and 9 physician assistants. Their mean age was 35 (⫾8), and 46% were female. The distribution of the children’s scores on the SAS was 0–17%; 1–9%; 2–30%; 3–14%; and 4–30%. Mean parental and practitioner scores on the VAS were 61 (⫾26) mm and 37 (⫾26) mm, respectively (maximal = 100). Correlation between child and parent scores was 0.47 (p < 0.001, Fig. 2). Correlation between child and practitioner scores was 0.08 (p = 0.54, Fig. 3). Correlation between parent and practitioner scores was 0.04 (p = 0.001).
DISCUSSION
Figure 1. Smiley Analog Scale (SAS). Reproduced with permission from: Pothman R. Comparison of the visual analog scale (VAS) and a smiley analog scale (SAS) for the evaluation of pain in children. Adv Pain Res Ther. 1990; 15:95–9.
It is now well recognized that many patients do not receive adequate analgesia while in the ED.12,13 Oligoanalgesia is particularly prominent in children,14–17 and there has been a generalized call for immediate correction of this ‘‘deplorable and un-
611
ACAD EMERG MED • June 2002, Vol. 9, No. 6 • www.aemj.org
Figure 2. Correlation between patient and parent pain scores.
Figure 3. Correlation between patient and practitioner pain scores.
ethical practice.’’3 An important prerequisite to reducing the phenomenon of oligoanalgesia in children is being able to recognize and quantify the severity of the child’s pain. However, research regarding pediatric pain assessment has been hampered by the misconceived notion that children have an altered perception of pain by their underdeveloped cognitive capacity.1 Some pediatric pain research studies have relied on parents or health care practitioners to determine the severity of pain in children, despite the fact that practitioners’ pain assessments do not correlate with their patients’ assessments, even in adults.11 It is generally agreed that self-reporting of pain is the most effective method of measuring pain.18 The VAS is the most commonly used tool to measure pain in adults. However, the abstract basis of the VAS may not be comprehended by young children. Recent studies have demonstrated that chil-
dren can reliably self-report pain severity using other age-appropriate scales once they reach age 4–5.19 A variety of abstract scales that have been used in children after they reach the concrete stage of development and include the Oucher, poker chip, and faces scales.3,6,19,8 The SAS is a pain rating scale that originated at the Pain Department of Sloan-Kettering Institute, New York, and was intended to better evaluate the emotional apperception of pain in children. This was adapted into a five-point scale by Pothman (Fig. 1).7 Using this scale, Pothman compared the pain ratings by VASs and the SAS in 96 children aged 3–18 years who were experiencing different painful disorders or procedures. In this study both scales were highly correlated (r = 0.87).7 In a subgroup of 23 children undergoing venipuncture, Pothman found no significant difference in the results as measured by the SAS between ages 3–9 and 10–18 years, and there was no reported failure using both scales. The results of the current study indicate that most children between the ages of 4 and 7 years can successfully use the SAS in the acute setting of the ED to indicate their pain severity. We also found that there is very poor correlation between the pain severity ratings of health care practitioners (using a VAS) and their pediatric patients. While the correlation between parental pain severity ratings and those of their children were better than for practitioners, they still were poor to fair. Not surprisingly, practitioner and parental pain ratings were also poorly correlated. These results suggest that both parental and health care practitioners are poor judges of their children’s/patient’s pain severity. The immediate implication of this study is that clinicians and researchers should not rely on their own judgment when making decisions regarding the assessment and management of pain in children. Instead they must ask the children to assess their own pain using any one of the many available age-appropriate scales. If the child is unable to assess his or her pain, clearly parental assessments are superior to those of the practitioners. Our results are not surprising in light of other pain-related studies. For example, Singer et al. compared the pain ratings of 1,104 patients undergoing 1,171 painful procedures with those of their practitioners and found that correlation between patient and practitioner pain scores for individual procedures was poor to fair (r = 0.26 to 0.68).11
LIMITATIONS Our study is limited by the small sample size. However, the 95% confidence intervals around the correlation coefficients imply poor to fair correla-
612
Singer et al. • JUDGMENT OF CHILDREN’S PAIN SEVERITY
tion between practitioner, parental, and patient pain scores. Also, we included a wide range of types of pain. It is possible that the results may have differed with a more homogeneous group. Our study results are limited to English-speaking, mostly white children, parents, and practitioners. As a result, it is unclear whether these results would generalize to other settings and populations. Another potential limitation is that it is often impossible to distinguish between pain and anxiety, particularly in children. Children were included in the study if the investigator thought that they were in pain at the time of the procedure. However, it is possible that some children were expressing their anxiety and not their pain. Finally, the Smiley scale cannot be considered the criterion standard for pain assessment in children. However, this6 and other similar studies7,8 have demonstrated construct validity when compared with other recognized scales. It is unclear whether there will ever be a true ‘‘gold standard’’ for measuring pain, given its subjective nature.
CONCLUSIONS The results of our study indicate that there is poor agreement between pain ratings by children, parents, and practitioners. While it is unclear which assessment best approximates the true degree of pain the child is experiencing, health care practitioners should not rely solely on their own or even parental assessments when managing pediatric patients with pain. Attempts to have the children assess their own pain severity should be made whenever possible. References 1.
Aradine CR, Beyer JE, Tompkins JM. Children’s pain perception before and after analgesia: a study of instrument construct validity and related issues. J Pediatr Nurs. 1988; 3:11–23.
2. 3.
4. 5. 6.
7.
8.
9.
10.
11.
12. 13. 14.
15.
16.
17.
18. 19.
Owen ME, Todt EH. Pain in infancy: neonatal reaction to a heel lance. Pain. 1984; 20:77–86. Williamson PS, Williamson PL. Physiologic stress reduction by a local anesthetic during circumcision. Pediatrics. 1983; 71:36–40. Terndrup TE. Pediatric pain control. Ann Emerg Med. 1996; 27:466–70. Ho K, Spence J, Murphy MF. Review of pain-measurement tools. Ann Emerg Med. 1996; 27:427–32. Hollander JE, Valentine SM, Brogan GX. The academic associate program: integrating clinical research with undergraduate education. Acad Emerg Med. 1997; 4:225–30. Pothman R. Comparison of the visual analog scale (VAS) and a smiley analog scale (SAS) for the evaluation of pain in children. Adv Pain Res Ther. 1990; 15:95–9. Maunukslea EL, Korpela R. Double-blind evaluation of a lognocaine-prilocaine cream (EMLA) in children. Br J Anaesth. 1986; 58:1242–5. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988. Karras DJ. Statistical methodology: II. Reliability and validity assessment in study design, Part A. Acad Emerg Med. 1997; 4:64–71. Singer AJ, Richman PB, Kowalska A, Thode HC. Comparison of patient and practitioner assessment of pain from commonly performed emergency department procedures. Ann Emerg Med. 1999; 33:652–8. Paris PM, Stewart (eds): Pain Management in Emergency Medicine. Norwalk, CT: Appleton & Lange, 1988. Wilson JE, Pendleton JM. Oligoanalgesia in the emergency department. Am J Emerg Med. 1987; 7:620–3. Friedland LR, Pancioli AM, Duncan KM. Pediatric emergency department analgesic practice. Pediatr Emerg Care. 1997; 13:103–6. Schechter NL, Allen DA, Hanson K. Status of pediatric pain control: a comparison of hospital analgesic usage in children and adults. Pediatrics. 1986; 77:11–5. Beyer JE, DeGood DE, Ashley LC, et al. Patterns of postoperative analgesic use with adults and children following cardiac surgery. Pain. 1983; 17:71–81. Petrack EM, Christopher NC, Kriwinsky J. Pain management in the emergency department: patterns of analgesic utilization. Pediatrics. 1997; 99:711–4. Beyer JE, Wells N. The assessment of pain in children. Pediatr Clin North Am. 1989; 36:837–54. McGrath PJ, de Veber LL, Heran MT. Multidimensional pain assessment in children. In: Fields HL (ed). Advances in Pain Research and Therapy, Vol 9. New York: Raven Press, 1986, pp 387–93.
