- Email: [email protected]
Discordance between patient self-reported visual analog scale pain scores and observed pain-related behavior in older children after surgery
ELSEVIER
Discordance Between Patient Self-Reported Visual Analog Scale Pain Scores and Observed Pain-Related Behavior in Older Children after Surgery Thomas R. Vetter, Department OH.
*Assistant Professor of Anesthesiology, Northeastern Ohio Universities College of Medicine, Rootstown OH; Director of Pediatric Pain Management tClinica1 Nurse Specialist for Pediatric Management
Pain
Address reprint requests to Dr. Vetter at the Department of Anesthesiology, Children’s Hospital Medical Center of Akron, One Perkins Square, Akron, OH 44308, USA. Received for publication June 29, 1995; revised manuscript accepted for publication October 30, 1995.
MD,*
of Anesthesiology,
Children’s
Elizabeth J. Heiner, Hospital
Medical
Center
MSNT of Akron,
Akron
Study Objective: To assess the correlation in an older pediatric population between, patient self-reported visual analog scale (VAS) p ain scores and observational pain-related behavior scores. Design: Prospective, comparative study. Setting: Inpatient surgical units of a free-standing children’s hospital. Patients: 30 ASA physical status I and II outpatients, 8 to 16 years of age, undergoing a van’ety of orthopedic, plastic, urologic, and general surgical procedures. Interventions: Each patient underwent a single assessment of pain intensity on thejrst postoperative day. Measurements and Main Results: Three health care providers (a clinical nurse specialist, a registered nurse with extensive pediatric experience, and a child life specialist) simultaneously generated an independent pain-related behavioral score {range of 0 = no pain to 100 = worst pain possible) based on their subjective perceptions of the patient’s observedfacial expression, activity level, and breathing pattern. The patient was then asked to provide a self-fepotied VAS pain score {range of 0 = ‘izo pain” with a smiling face, to 100 = “worst pain ever” with a frowning face). A single set of such concurrent pain scores was obtained once from each study patient. The pain-related behavior scores displayed a satisfactovy interrater reliability, with an observed interclass correlation coefficient (Cohen’s kappa value) of 0.83. When compared with a patient’s selfreported VAS score, the three pain-related behavioral scores generated by each health care provider for a given patient exhibited both variable and minimal correlation. Conclusions: A tenuous relationship may exist between an older child’s own perception of pain intensity and his or her behavioral expression of that pain as intupreted by a health care provider.
Keywords: Pain: measurement behavior; visual analog scale.
of, pediatric,
postoperative;
pediatrics:
Introduction In light of the fundamental research and clinical requirements for accurate pain assessment, within the last two decades an enormous amount of energy has been
Journal of Clinical Anesthesia 8:371-375, 1996 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0952-8180/96/$15.00 PI1 SO952-8180(96)00079-7
Ori@nal Contn’butions
devoted to developing techniques to assess and to quantify pediatric pain intensity. The end result has been an array of behavioral, physiologic, and psychological scaling tools with varying degrees of complexity and attendant clinical applicability.‘-3 Due in large part to their simplicity and hence utility in a busy clinical setting, so-called unidimensional pain scales, such as the visual analog scale (VAS), graphic rating, and numerical rating scales, have seen widespread application in school-age children. However, while intended strictly to measure pain intensity, these unidimensional patient self-report scales also reflect the effects of a complex summation of cognitive, personality, learning, cultural, and environmental influences on an individual child’s pain perception.4 Moreover, it has been demonstrated that a significant discordance exists in children three to seven years of age between concurrent patient self-reported pain scores and observational pain-related behavior scores generated by health care providers.5 The present study was undertaken to assess the correlation in an older pediatric population between patient self-reported VAS scores and observational pain-related behavior scores and, hence, to determine the usefulness of such pain-related behavioral observations as one component of a proposed multifactorial pediatric pain scoring system.
Materials
and Methods
This study was approved by the Institutional Review Board of the Children’s Hospital Medical Center of Akron. After written, informed parental consent was obtained, 30 ASA physical status I and II outpatients, 8 to 16 years of age, undergoing a variety of orthopedic, plastic, urologic, and general surgical procedures that required same day admission postoperatively, were enrolled in the study. Exclusionary criteria included mental retardation, cerebral palsy, and any other medical condition that would seemingly unduly affect the collection of conventional pain scale scores. All of the enrolled patients received standardized, preoperative instruction on the use of the 10 cm slide-rule type of VAS (Figure 1) to be used for postoperative data collection. In a consistent manner on the morning of the first postoperative day, with at least one of the study patient’s parents present in the child’s hospital room, three health care providers (a pediatric clinical nurse specialist for pain management; a registered nurse with extensive pediatric experience; and a child life specialist whose primary role is to reduce any adverse psychological impact from the child’s hospitalization) simultaneously generated an independent pain-related behavioral score (range 0 = no pain to 100 = worst pain possible) based on their subjective perceptions of the patient’s observed facial expression, activity level, and breathing pattern (Figure 2). Each of the health care providers had been instructed at the outset of the study to generate an overall clinical impression of the child’s pain intensity, as they would in their daily clinical practice, using the three pain-related behav-
372
J. Clin. Anesth.,
vol. 8, August
1996
Figure 1. Visual analog scale slide ruler used for the measurement of postoperative pain intensity. The graphic scale (above) was used initially by the study patient to report pain intensity. The numerical scale (shown below) on the reverse side of the slide ruler was then used to determine the corresponding numerical pain score value (0 to 100). iors. The patient was then asked to provide a self-reported VAS pain score (range 0 = “no pain” with a smiling face to 100 = “worst pain ever” with a frowning face). A single set of such concurrent pain scores was obtained once from each study patient. The pain-related behavioral scores simultaneously generated for each study patient by the three observers were subjected to interclass correlation analysis (kappa) in order to assess interrater reliability. The three independent pain-related behavioral scores for a given patient were then compared using Pearson product-moment correlation analysis with that patient’s own concurrent selfreported VAS score. Pearson product-moment correlation analysis was applicable to the two types of pain scores obtained due to their being both continuous variables and independent measures across time.
Results Complete data, comprised of three independent painrelated behavioral scores and one VAS score for each patient, were collected from a total of 30 patients. The painrelated behavioral scores displayed a satisfactory interrater
Observer:
Based upon observations related
by healthcare
provider of three pain-
behaviors:
1) Facial Expression 2) Activity
Level
3) Breathing
Pattern
Rate the patient’s pain between 0 = no pain and 100 = worst pain possible.
Heelthcare
Figure
Provider’s
Pain Score:
2. Scale for observation of pain-related
behaviors.
reliability with an observed interclass correlation coeffcient (Cohen’s kappa value) of 0.83. When compared with a patient’s self-reported VAS score, the three pain-related behavioral scores generated by each health care provider for a given patient exhibited both minimal Pearson correlation coefficients 0.33, 0.65, and 0.66, and a very random plot pattern (Fig-Llre3).
Discussion Enormous progress has been made within the last 15 years in improving both the global awareness of the need for and the ability to provide effective pediatric pain manage-
ment.6 As the field of pediatric pain management continues to advance, pediatric pain assessment appears to be one fundamental area in need of further investigation and refinement? A reevaluation of our present understanding of pediatric pain assessment is necessary, because significant questions remain regarding the accuracy and validity of many of the observational, self-reported, and physiologic pediatric pain measurement scales that have been developed and widely applied both clinically and in research design.* As an example, previous authors have described a significant discordance between concurrent patient self-reported pain scores and observational painrelated behavior scores in children three to seven years of
100
0
100:
Observer B
Observer A
.
.
90
.
.
.
.
60
60-
50 40 30
.
20 10
Y = 0.665(X)+10.55: 0
10
20
30
40
50
60
70
80
90
B
Patient Self-Reported Visual Analog Pain Score
A
ObwNer
30
50
R”2=0.122
0
10
20 Patient
40
Self-ReportedVlbuel
60 Analog
70
a0
90
100
Pain Score
C .
.
go-
.
. .
.
80: 7060-
‘5 d2
507
‘FI g
40.
% f8
30:
B $
20.
a 0
lo-
l
.
l .
.
.
o---Y=O.ffiZ(X)+ -lOl---10
C
i21.46;
.
100:
e 8 w .t 2 8
Y= 0.332(X) / 7-
-10
100
.
.
Fl”2 = 0.462 -10
-10
.
-
0
0
10
20
30
40
14.8; I+‘2 = 0.383 50
60
70
80
Patient Self-Repotted Visual Analog Pain Score
90
100
Figure 3. Regression analyses for the patient’s own visual analog scale pain scores versus the health care provider’s observed pain-related behavioral pain score. A. Observer A: pediatric clinical nurse specialist. B. Observer B: pediatric registered nurse. C. Observer C: child life specialist.
J. Clin. Anesth., vol. 8, August 1996
373
Original Contributions
age, with a common occurrence being minimal behavioral manifestations despite self-reports of severe pain by the child.5 The primary motivation for undertaking the present study was to delineate whether older children display pain behaviors that are more consistent with their own perceived pain intensity as compared with their younger pediatric counterparts. In the process, insight also was sought as to the merit of including such clinical behavioral observations as one component of a multifactorial pediatric pain assessment. One major component in the training of any clinician involves the development of observational skills and the ability to diagnose based on the data so acquired. Clinicians thus come to rely heavily on their direct observation of a patient in order to draw a clinical conclusion. This phenomenon extends to the assessment of pain intensity via the observation of pain-related behaviors in both the adult and child. Three pain-related behaviors frequently observed in clinical practice include facial expression, activity level, and breathing pattern9 As was done by the three health care providers in the present study, it is common clinical practice to take note of a select number of pain-related behaviors and subsequently generate a subjective impression of the patient’s immediate pain intensity. Hence, the need for and efficacy of analgesic administrati0n.l’ Such a global rating of pain intensity by independent observers has been shown to correlate very well with more detailed behavioral methods and has been applied in previously reported studies of postoperative pediatric pain.r”’ Nonetheless, based on the data presented here, it would appear that health care professionals should avoid placing undue emphasis on their clinical observations in assessing postoperative pain intensity in children 8 to 16 years of age. The VAS has been in widespread use in the literature and achieved clinical preeminence as an adult and pediatric pain assessment tool. “a13 However, the VAS may often have a shortcoming in that despite being intended to measure strictly pain intensity, it also reflects the effects of diverse predisposing and situational psychosocial contributors to a child’s acute pain experience.4214 TO remedy this shortcoming to presently accepted methods, “a comprehensive approach for pain assessment in children requires an evaluation of the pain complaint (the sensory dimension) in relation to the child experiencing it (emotional, familial, and situational factors) .“3 Unfortunately, many clinicians with a strong basic science orientation, including anesthesiologists, tend to reject such psychosocial theory out of hand, which may effectively hamper their accurate clinical interpretation of a pediatric patient’s VAS pain scores. In response to the complexity of the pain experience, multifactorial pediatric pain assessment has been developed that incorporates various behavioral, physiologic, and/or self-administered patient-controlled analgesia (PCA) dosage subscales in order to confirm patient selfreports of acute pain intensity. The primary rationale for incorporating such subscales has been to increase the va-
374
J. Clin. Anesth., vol. 8, August 1996
lidity of the pain assessment effort. Similarly, such multidimensional pain assessment is innately attractive because it mimics one’s actual clinical practice of creating an impression of a child’s pain intensity based on several parameters. However, such assessment may have little merit unless it takes into consideration and somehow quantitates the effects of critical underlying psychosocial factors. In summary, based on the data presented here, it would appear that a tenuous relationship may exist between an older child’s own perception of pain intensity and his or her behavioral expression of that pain as interpreted by a health care provider. One of the challenges remaining in acute pediatric pain management is the development of a simplified yet valid multidimensional pain assessment tool that addresses the entire postoperative pediatric pain experience.
Acknowledgments The authors would like to thank Karen Addis, RN, and Gena Valloric, CCLS, for their invaluable assistance in collecting the observational pains scale data reported in this study.
References McGrath PJ, Cunningham CJ, Goodman JT, Unruh AM: The clinical measurement of pain in children: a review. Clin J Pain 1986;2:221-7. McGrath PA, de Veber LL, Hearn MT: Multidimensional pain assessment in children. In: Fields HL, Dubner R, Cervero F (eds): Advances in Pain Research and Therapy, vol. 9. New York: Raven Press, 1985:387-93. McGrath PA: An assessment of children’s pain: a review of behavioral, physiological and direct scaling techniques. Pain 1987; 31:147-76. Chapman CR: Measurement of pain: problems and issues. In: Bonicag (ed): Advances in Pain Research and Therapy, vol. 1. New York: Raven Press, 1976:345-54. BeyerJE, McGratb PJ, Berde CB: Discordance between self-report and behavioral pain measures in children aged 3-7 years after surgery. J Pain Symptom Manage 1990;5:350-6. Houck CS, Berde CB, Anand KJS: Pediatric pain management. In: Gregory GA (ed): Pediatric Anesthesia, 3rd ed. New York: Churchill Livingstone, 1994,743-71. 7. Tobias JD: Pediatric pain management: perspectives on the past, discussions of the present, and speculations on the future. AmJ Pai% Manage 1991;1:12-5. 8. Erickson CJ: Pain measurement in children: tions. JDm Behav Pediatr 1990;11:128-34.
problems and direc-
9. Maunuksela EL, Olkkola KT, Korpela R: Measurement of pain in children with self-reporting and behavioral assessment. Clin Pharmacol The-r 1987;42:137-41. 10. McGrath PJ, Unruh AM: Pain in adolescents and children. Amsterdam: Elsevier, 1987:73-104.
Pain scores in older children: Vetter and Heiner 11. Martin LV: Postoperative analgesia after circumcision in children. BrJAnaesth 1982;54:1263-6. 12. Dexter F, Chestnut DH: Analysis of statistical tests to compare visual analog scale measurements among groups. Anesthesiology 1995;82:896-902. 13. Mantha S, Thisted R, Foss J, Ellis JE, Roizen MF: A proposal to use
Discordance
confidence intervals for visual analog scale data for pain measurement to determine clinical significance. An& Analg 1993; 77:1041-7. 14. Chapman CR Psychological factors in postoperative pain. In: Smith G, Covino BG (eds.): Acute Pain. London: Butter-worth% 1985:22-41.
Between Self-Report and Behavioral Pain Measures in Children Aged 3-7 Years after Surgery Department
J.E. Beyer, PJ of Anesthesia,
McGrath, Children’s
C.B. Berde Hospital, Boston, MA
Abstract This study examined concurrent self-reports of pain intensity and behavioral responses in 25 children aged 3-7 yr. Behavioral (Children’s Hospital of Eastern Ontario Pain Scale, CHEOPS) and self-report (the Oucher and Analogue Chromatic Continuous Scale) measures of pain were obtained following major surgery. The two self-report measures were strongly and significantly correlated, and the pattern of scores over the 36hr observation period was as expected. There was little relationship between the scores for the self-report and the behavioral measures. Many children who reported severe pain manifested few of the behavioral indicators of distress used in the CHEOPS. This behavioral response pattern may occur commonly in children experiencing pain after surgery and may limit the applicability of current behavioral scales as sole measures of pain intensity in younger children. Reprinted
from Journal of Pain and Symptom Management 1990;5:350-6.
J. Clin. Anesth., vol. 8, August 1996
375
Discordance Between Patient Self-Reported Visual Analog Scale Pain Scores and Observed Pain-Related Behavior in Older Children after Surgery Thomas R. Vetter, Department OH.
*Assistant Professor of Anesthesiology, Northeastern Ohio Universities College of Medicine, Rootstown OH; Director of Pediatric Pain Management tClinica1 Nurse Specialist for Pediatric Management
Pain
Address reprint requests to Dr. Vetter at the Department of Anesthesiology, Children’s Hospital Medical Center of Akron, One Perkins Square, Akron, OH 44308, USA. Received for publication June 29, 1995; revised manuscript accepted for publication October 30, 1995.
MD,*
of Anesthesiology,
Children’s
Elizabeth J. Heiner, Hospital
Medical
Center
MSNT of Akron,
Akron
Study Objective: To assess the correlation in an older pediatric population between, patient self-reported visual analog scale (VAS) p ain scores and observational pain-related behavior scores. Design: Prospective, comparative study. Setting: Inpatient surgical units of a free-standing children’s hospital. Patients: 30 ASA physical status I and II outpatients, 8 to 16 years of age, undergoing a van’ety of orthopedic, plastic, urologic, and general surgical procedures. Interventions: Each patient underwent a single assessment of pain intensity on thejrst postoperative day. Measurements and Main Results: Three health care providers (a clinical nurse specialist, a registered nurse with extensive pediatric experience, and a child life specialist) simultaneously generated an independent pain-related behavioral score {range of 0 = no pain to 100 = worst pain possible) based on their subjective perceptions of the patient’s observedfacial expression, activity level, and breathing pattern. The patient was then asked to provide a self-fepotied VAS pain score {range of 0 = ‘izo pain” with a smiling face, to 100 = “worst pain ever” with a frowning face). A single set of such concurrent pain scores was obtained once from each study patient. The pain-related behavior scores displayed a satisfactovy interrater reliability, with an observed interclass correlation coefficient (Cohen’s kappa value) of 0.83. When compared with a patient’s selfreported VAS score, the three pain-related behavioral scores generated by each health care provider for a given patient exhibited both variable and minimal correlation. Conclusions: A tenuous relationship may exist between an older child’s own perception of pain intensity and his or her behavioral expression of that pain as intupreted by a health care provider.
Keywords: Pain: measurement behavior; visual analog scale.
of, pediatric,
postoperative;
pediatrics:
Introduction In light of the fundamental research and clinical requirements for accurate pain assessment, within the last two decades an enormous amount of energy has been
Journal of Clinical Anesthesia 8:371-375, 1996 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0952-8180/96/$15.00 PI1 SO952-8180(96)00079-7
Ori@nal Contn’butions
devoted to developing techniques to assess and to quantify pediatric pain intensity. The end result has been an array of behavioral, physiologic, and psychological scaling tools with varying degrees of complexity and attendant clinical applicability.‘-3 Due in large part to their simplicity and hence utility in a busy clinical setting, so-called unidimensional pain scales, such as the visual analog scale (VAS), graphic rating, and numerical rating scales, have seen widespread application in school-age children. However, while intended strictly to measure pain intensity, these unidimensional patient self-report scales also reflect the effects of a complex summation of cognitive, personality, learning, cultural, and environmental influences on an individual child’s pain perception.4 Moreover, it has been demonstrated that a significant discordance exists in children three to seven years of age between concurrent patient self-reported pain scores and observational pain-related behavior scores generated by health care providers.5 The present study was undertaken to assess the correlation in an older pediatric population between patient self-reported VAS scores and observational pain-related behavior scores and, hence, to determine the usefulness of such pain-related behavioral observations as one component of a proposed multifactorial pediatric pain scoring system.
Materials
and Methods
This study was approved by the Institutional Review Board of the Children’s Hospital Medical Center of Akron. After written, informed parental consent was obtained, 30 ASA physical status I and II outpatients, 8 to 16 years of age, undergoing a variety of orthopedic, plastic, urologic, and general surgical procedures that required same day admission postoperatively, were enrolled in the study. Exclusionary criteria included mental retardation, cerebral palsy, and any other medical condition that would seemingly unduly affect the collection of conventional pain scale scores. All of the enrolled patients received standardized, preoperative instruction on the use of the 10 cm slide-rule type of VAS (Figure 1) to be used for postoperative data collection. In a consistent manner on the morning of the first postoperative day, with at least one of the study patient’s parents present in the child’s hospital room, three health care providers (a pediatric clinical nurse specialist for pain management; a registered nurse with extensive pediatric experience; and a child life specialist whose primary role is to reduce any adverse psychological impact from the child’s hospitalization) simultaneously generated an independent pain-related behavioral score (range 0 = no pain to 100 = worst pain possible) based on their subjective perceptions of the patient’s observed facial expression, activity level, and breathing pattern (Figure 2). Each of the health care providers had been instructed at the outset of the study to generate an overall clinical impression of the child’s pain intensity, as they would in their daily clinical practice, using the three pain-related behav-
372
J. Clin. Anesth.,
vol. 8, August
1996
Figure 1. Visual analog scale slide ruler used for the measurement of postoperative pain intensity. The graphic scale (above) was used initially by the study patient to report pain intensity. The numerical scale (shown below) on the reverse side of the slide ruler was then used to determine the corresponding numerical pain score value (0 to 100). iors. The patient was then asked to provide a self-reported VAS pain score (range 0 = “no pain” with a smiling face to 100 = “worst pain ever” with a frowning face). A single set of such concurrent pain scores was obtained once from each study patient. The pain-related behavioral scores simultaneously generated for each study patient by the three observers were subjected to interclass correlation analysis (kappa) in order to assess interrater reliability. The three independent pain-related behavioral scores for a given patient were then compared using Pearson product-moment correlation analysis with that patient’s own concurrent selfreported VAS score. Pearson product-moment correlation analysis was applicable to the two types of pain scores obtained due to their being both continuous variables and independent measures across time.
Results Complete data, comprised of three independent painrelated behavioral scores and one VAS score for each patient, were collected from a total of 30 patients. The painrelated behavioral scores displayed a satisfactory interrater
Observer:
Based upon observations related
by healthcare
provider of three pain-
behaviors:
1) Facial Expression 2) Activity
Level
3) Breathing
Pattern
Rate the patient’s pain between 0 = no pain and 100 = worst pain possible.
Heelthcare
Figure
Provider’s
Pain Score:
2. Scale for observation of pain-related
behaviors.
reliability with an observed interclass correlation coeffcient (Cohen’s kappa value) of 0.83. When compared with a patient’s self-reported VAS score, the three pain-related behavioral scores generated by each health care provider for a given patient exhibited both minimal Pearson correlation coefficients 0.33, 0.65, and 0.66, and a very random plot pattern (Fig-Llre3).
Discussion Enormous progress has been made within the last 15 years in improving both the global awareness of the need for and the ability to provide effective pediatric pain manage-
ment.6 As the field of pediatric pain management continues to advance, pediatric pain assessment appears to be one fundamental area in need of further investigation and refinement? A reevaluation of our present understanding of pediatric pain assessment is necessary, because significant questions remain regarding the accuracy and validity of many of the observational, self-reported, and physiologic pediatric pain measurement scales that have been developed and widely applied both clinically and in research design.* As an example, previous authors have described a significant discordance between concurrent patient self-reported pain scores and observational painrelated behavior scores in children three to seven years of
100
0
100:
Observer B
Observer A
.
.
90
.
.
.
.
60
60-
50 40 30
.
20 10
Y = 0.665(X)+10.55: 0
10
20
30
40
50
60
70
80
90
B
Patient Self-Reported Visual Analog Pain Score
A
ObwNer
30
50
R”2=0.122
0
10
20 Patient
40
Self-ReportedVlbuel
60 Analog
70
a0
90
100
Pain Score
C .
.
go-
.
. .
.
80: 7060-
‘5 d2
507
‘FI g
40.
% f8
30:
B $
20.
a 0
lo-
l
.
l .
.
.
o---Y=O.ffiZ(X)+ -lOl---10
C
i21.46;
.
100:
e 8 w .t 2 8
Y= 0.332(X) / 7-
-10
100
.
.
Fl”2 = 0.462 -10
-10
.
-
0
0
10
20
30
40
14.8; I+‘2 = 0.383 50
60
70
80
Patient Self-Repotted Visual Analog Pain Score
90
100
Figure 3. Regression analyses for the patient’s own visual analog scale pain scores versus the health care provider’s observed pain-related behavioral pain score. A. Observer A: pediatric clinical nurse specialist. B. Observer B: pediatric registered nurse. C. Observer C: child life specialist.
J. Clin. Anesth., vol. 8, August 1996
373
Original Contributions
age, with a common occurrence being minimal behavioral manifestations despite self-reports of severe pain by the child.5 The primary motivation for undertaking the present study was to delineate whether older children display pain behaviors that are more consistent with their own perceived pain intensity as compared with their younger pediatric counterparts. In the process, insight also was sought as to the merit of including such clinical behavioral observations as one component of a multifactorial pediatric pain assessment. One major component in the training of any clinician involves the development of observational skills and the ability to diagnose based on the data so acquired. Clinicians thus come to rely heavily on their direct observation of a patient in order to draw a clinical conclusion. This phenomenon extends to the assessment of pain intensity via the observation of pain-related behaviors in both the adult and child. Three pain-related behaviors frequently observed in clinical practice include facial expression, activity level, and breathing pattern9 As was done by the three health care providers in the present study, it is common clinical practice to take note of a select number of pain-related behaviors and subsequently generate a subjective impression of the patient’s immediate pain intensity. Hence, the need for and efficacy of analgesic administrati0n.l’ Such a global rating of pain intensity by independent observers has been shown to correlate very well with more detailed behavioral methods and has been applied in previously reported studies of postoperative pediatric pain.r”’ Nonetheless, based on the data presented here, it would appear that health care professionals should avoid placing undue emphasis on their clinical observations in assessing postoperative pain intensity in children 8 to 16 years of age. The VAS has been in widespread use in the literature and achieved clinical preeminence as an adult and pediatric pain assessment tool. “a13 However, the VAS may often have a shortcoming in that despite being intended to measure strictly pain intensity, it also reflects the effects of diverse predisposing and situational psychosocial contributors to a child’s acute pain experience.4214 TO remedy this shortcoming to presently accepted methods, “a comprehensive approach for pain assessment in children requires an evaluation of the pain complaint (the sensory dimension) in relation to the child experiencing it (emotional, familial, and situational factors) .“3 Unfortunately, many clinicians with a strong basic science orientation, including anesthesiologists, tend to reject such psychosocial theory out of hand, which may effectively hamper their accurate clinical interpretation of a pediatric patient’s VAS pain scores. In response to the complexity of the pain experience, multifactorial pediatric pain assessment has been developed that incorporates various behavioral, physiologic, and/or self-administered patient-controlled analgesia (PCA) dosage subscales in order to confirm patient selfreports of acute pain intensity. The primary rationale for incorporating such subscales has been to increase the va-
374
J. Clin. Anesth., vol. 8, August 1996
lidity of the pain assessment effort. Similarly, such multidimensional pain assessment is innately attractive because it mimics one’s actual clinical practice of creating an impression of a child’s pain intensity based on several parameters. However, such assessment may have little merit unless it takes into consideration and somehow quantitates the effects of critical underlying psychosocial factors. In summary, based on the data presented here, it would appear that a tenuous relationship may exist between an older child’s own perception of pain intensity and his or her behavioral expression of that pain as interpreted by a health care provider. One of the challenges remaining in acute pediatric pain management is the development of a simplified yet valid multidimensional pain assessment tool that addresses the entire postoperative pediatric pain experience.
Acknowledgments The authors would like to thank Karen Addis, RN, and Gena Valloric, CCLS, for their invaluable assistance in collecting the observational pains scale data reported in this study.
References McGrath PJ, Cunningham CJ, Goodman JT, Unruh AM: The clinical measurement of pain in children: a review. Clin J Pain 1986;2:221-7. McGrath PA, de Veber LL, Hearn MT: Multidimensional pain assessment in children. In: Fields HL, Dubner R, Cervero F (eds): Advances in Pain Research and Therapy, vol. 9. New York: Raven Press, 1985:387-93. McGrath PA: An assessment of children’s pain: a review of behavioral, physiological and direct scaling techniques. Pain 1987; 31:147-76. Chapman CR: Measurement of pain: problems and issues. In: Bonicag (ed): Advances in Pain Research and Therapy, vol. 1. New York: Raven Press, 1976:345-54. BeyerJE, McGratb PJ, Berde CB: Discordance between self-report and behavioral pain measures in children aged 3-7 years after surgery. J Pain Symptom Manage 1990;5:350-6. Houck CS, Berde CB, Anand KJS: Pediatric pain management. In: Gregory GA (ed): Pediatric Anesthesia, 3rd ed. New York: Churchill Livingstone, 1994,743-71. 7. Tobias JD: Pediatric pain management: perspectives on the past, discussions of the present, and speculations on the future. AmJ Pai% Manage 1991;1:12-5. 8. Erickson CJ: Pain measurement in children: tions. JDm Behav Pediatr 1990;11:128-34.
problems and direc-
9. Maunuksela EL, Olkkola KT, Korpela R: Measurement of pain in children with self-reporting and behavioral assessment. Clin Pharmacol The-r 1987;42:137-41. 10. McGrath PJ, Unruh AM: Pain in adolescents and children. Amsterdam: Elsevier, 1987:73-104.
Pain scores in older children: Vetter and Heiner 11. Martin LV: Postoperative analgesia after circumcision in children. BrJAnaesth 1982;54:1263-6. 12. Dexter F, Chestnut DH: Analysis of statistical tests to compare visual analog scale measurements among groups. Anesthesiology 1995;82:896-902. 13. Mantha S, Thisted R, Foss J, Ellis JE, Roizen MF: A proposal to use
Discordance
confidence intervals for visual analog scale data for pain measurement to determine clinical significance. An& Analg 1993; 77:1041-7. 14. Chapman CR Psychological factors in postoperative pain. In: Smith G, Covino BG (eds.): Acute Pain. London: Butter-worth% 1985:22-41.
Between Self-Report and Behavioral Pain Measures in Children Aged 3-7 Years after Surgery Department
J.E. Beyer, PJ of Anesthesia,
McGrath, Children’s
C.B. Berde Hospital, Boston, MA
Abstract This study examined concurrent self-reports of pain intensity and behavioral responses in 25 children aged 3-7 yr. Behavioral (Children’s Hospital of Eastern Ontario Pain Scale, CHEOPS) and self-report (the Oucher and Analogue Chromatic Continuous Scale) measures of pain were obtained following major surgery. The two self-report measures were strongly and significantly correlated, and the pattern of scores over the 36hr observation period was as expected. There was little relationship between the scores for the self-report and the behavioral measures. Many children who reported severe pain manifested few of the behavioral indicators of distress used in the CHEOPS. This behavioral response pattern may occur commonly in children experiencing pain after surgery and may limit the applicability of current behavioral scales as sole measures of pain intensity in younger children. Reprinted
from Journal of Pain and Symptom Management 1990;5:350-6.
J. Clin. Anesth., vol. 8, August 1996
375