American Journal of Emergency Medicine (2007) 25, 739 - 742

Original Contribution

Clinically significant changes in acute pain in a pediatric ED using the Color Analog Scale

Tom McConahay MD^a, Matt Bryson BSc^b, Blake Bulloch MD^a,*

^aDepartment of Emergency Medicine, Phoenix Children’s Hospital, Phoenix, AZ 85016, USA

^bDepartment of Child Life, Phoenix Children’s Hospital, Phoenix, AZ 85016, USA

Received 27 October 2006; revised 1 December 2006; accepted 1 December 2006

Abstract

Objective: The purpose of this study is to quantify, using the Color Analog Scale (CAS), the degree of change in Pain severity required to achieve a clinically significant improvement in pain.

Methods: A prospective descriptive study, using convenience sampling of children aged 5 and 12 years presenting to a pediatric emergency department (ED) with acute pain, was done. Children were asked to mark their pain severity on a previously validated CAS. After a pain intervention, the child was again asked to mark their pain intensity on the CAS and asked to describe the relative change in their pain. The main outcome measure was to quantify the smallest change required for the child to state that their pain was improved.

Results: One hundred twenty-six children with a mean age of 8.6 years (SD, 2.8 years) were enrolled. Males accounted for 56%. Pain was traumatic in 47.6% and nontraumatic in 52.4%. Of the 126 pain comparisons made, 28 children described their pain as bthe sameQ and had a mean change in score of

-0.10 cm (95% confidence interval [CI], -2.27 to 2.07 cm). Pain was judged to be a blittle lessQ in 58 children, and the CAS score changed by a mean of -2.4 cm (95% CI, -3.15 to -1.72 cm). In the 29 children who judged their pain to be bmuch less,Q the CAS score decreased by a mean of -5.4 cm (95% CI, -6.50 to -4.40).

Conclusion: A decrease on the CAS pain scale of 2.4 cm (95% CI, 2.95-1.92) is a clinically significant change in pain for children aged 5 to 12 years with acute pain. The CAS is a valuable tool in assessing responses to pain interventions.

D 2007

Introduction

The assessment and management of pain, especially in the emergency department (ED), is extremely important.

This study was supported by an unrestricted investigator-initiated study grant provided by McNeil Pharmaceuticals, Titusville, NJ.

This study was presented as a poster presentation at the Pediatric Ambulatory Societies Annual Meeting, April 2005, in Washington, DC.

* Corresponding author. Tel.: +602 546 1910; fax: +602 546 1907.

E-mail address: [email protected] (B. Bulloch).

The evaluation and treatment of pain among children is becoming an increasingly important aspect of emergency care. In 2001, The Joint Commission on Accreditation of Health Care Organizations mandated the assessment and documentation of pain for all health care encounters [1]. There have been numerous reports of the inadequacy of pain recognition and relief. Recently, Drendel et al [2] showed that only 44.5% of ED visits had documented pain scores, and this often leads to the undertreatment of pain in the ill or injured child.

0735-6757/$ - see front matter D 2007 doi:10.1016/j.ajem.2006.12.010

Self-report is recognized as being the most effective method to measure pain, with children as young as 3 years old being able to reliably self-report pain intensity [3]. Adequate assessment of pain requires a tool that has been validated in the setting in which it will be used. Another important aspect in pain evaluation is to determine the minimum clinically significant difference in pain experi- ence, thus, allowing for documentation of pain relief after an intervention [4].

There are many pain scales available that can be used for pain assessments in a variety of settings. However, the Color Analog Scale (CAS) is one of the only pain scales that has been validated for use in children 5 years and older with acute pain who present to the ED [5]. In addition, the CAS is easy to use, inexpensive, and fast to administer, making it ideal in a busy ED setting. Only one previous study has examined what constitutes a significant change in pain scores using the CAS [6]. As such, reexamining the scale in a different population is warranted to ensure the generaliz- ability of the results.

The objective of our study was to determine the clinically significant change in pain scale scores using the CAS in children between the ages of 5 and 12 years presenting to a pediatric ED.

Methods

We performed a prospective descriptive study using convenience sampling in the ED of an urban Children’s hospital with an annual census of 55000. The study was approved by the site’s institutional review board. From September 2003 to July 2004, all children presenting to the ED aged 5 to 12 years were eligible for enrollment if they had a complaint of pain. They were enrolled when a research assistant was available, mainly between 1500 and 2400 hours, corresponding with the peak Daily census.

Children were excluded if they had an Altered level of consciousness, were clinically unstable, did not speak English or Spanish, or were developmentally delayed. Research assistants approached all eligible children for verbal assent and their parents or legal guardians for written informed consent. Children were asked to mark their current pain severity on a previously validated standardized 10-cm CAS anchored by the descriptors bno painQ and bmost pain.Q The CAS is 20 mm wide at the bottom and white in color, reflecting no pain. It widens to 75 mm at the top and gradually changes in color to red indicating the worst pain. The reverse side of the scale is numbered from 0 to 10 in 1-cm increments, enabling investigators to quantify the children’s responses. The exact method by which children were instructed in its use was by script as done previously by Bulloch and Tenenbein [5,6]. bSee this scale, it is like a thermometer. The bottom, where it is small and white means no pain and the top where it is wider and red means the most pain you can imagine (research assistant pointed to each end

while describing). You slide the marker to the spot on the thermometer that shows how much pain you are having right now.Q All measurements were recorded to the closest

0.25 cm.

At this time, children were also asked if they considered their pain to be mild, moderate, or severe. If they did not understand these terms, they were asked if it hurt ba little,Q ba lot,Q or bsomewhere in between.Q The purpose of this was to ensure we had encompassed the entire pain spectrum.

After their initial pain assessment, a variety of pain interventions were conducted based on standard clinical practice. The pain interventions were at the discretion of the attending physician and included fracture immobilization, intravenous, oral, or topical analgesics. At 30 minutes, all children had their pain score reassessed by the same research assistant using the same protocol as before. All children were blinded to their previous scores. At this time, they were also asked about the change in their pain intensity if any. Precisely, they were asked if their pain was bmuch less,Q a blittle less,Q bthe same,Q a blittle worse,Q or bmuch worse.Q This is a Likert scale that has only been validated in children as young as 5 years [7]. It is for this reason that children younger than 5 years were excluded from the study. The key outcome measure was the smallest change on the CAS (measured to the closest 0.25 cm), demonstrated quantitatively with 95% confidence intervals (CIs) neces- sary to cause the child to note the pain to be a blittle less.Q This was defined as the clinically significant change in pain. The bidealQ change in pain was defined as the amount of change necessary for the child to claim that the pain was resolved, bmuch lessQ or when they felt they no longer

required anything to help the pain go away [6].

Data were entered into a spreadsheet program. Analysis was performed by a trained biomedical statistician with the use of a statistical software package (SAS, Cary, NC). Simple descriptive statistics were used to describe the characteristics of the sample. Point and interval estimates of effect were used to determine the differences in pain scores before and after analgesic administration.

Results

A total of 126 children with a mean age of 8.6 years (SD, 2.8 years; range, 5-12 years) were enrolled. Seventy-one (56.3%) were male; 65 (51.6%), Hispanic; 47 (37.3%),

Caucasian; 4 (3.2%), African American; and 10 (7.9%) fell into the bothersQ category. Overall, on initial assessment, 24 (19%) children considered their pain to be mild, 48 (38%) considered their pain to be moderate, and 54 (43%) considered their pain to be severe by self-report.

Trauma was the etiology of the pain in 60 (47.6%) cases and consisted of 52 (86.6%) fractures and 8 (13.4%) soft tissue injuries. Nontraumatic conditions accounted for 66 (52.4%) cases and included mainly children with abdominal pain (50%) and headaches (24%).

Compared with their initial pain assessment, 28 children described their pain as bthe sameQ and had a mean change in score of -0.10 cm (95% CI, -2.27 to 2.07). Of the 58 pain comparisons that were judged to be a blittle less,Q the CAS score changed by a mean of -2.40 cm (95% CI, -3.15 to

-1.72). In the 29 children who judged their pain to be

bmuch less,Q the CAS score decreased by a mean of

-5.40 cm (95% CI, -6.50 to -4.40). Only 10 pain

comparisons were judged to be a blittle worse,Q and their score changed by a mean of 0.9 cm (95% CI, 0.10 to 1.70). In the one child who judged their pain to be bmuch worse,Q the CAS score increased by 0 cm.

Discussion

Several pain scales that can be used to assess changes in pain severity over time exist. A common one is the Visual analogue scale (VAS). Investigators in the adult popula- tion have quantified a reduction of 6.2 to 23 mm as clinically significant using the VAS [4,8-12]. Powell et al

[8] used a nonvalidated 100-mm nonhatched VAS in children between 8 and 15 years old and determined the minimum clinically significant change in VAS score to be 10 mm (95% CI, 7-12 mm).

The CAS has not only been validated for the pediatric population in an emergency setting, but a previous study attempted to define the change needed for clinically significant improvement in pain [5,6]. They found a mean decrease of 1.7 cm to be a clinically significant decrease in pain and a mean decrease of 3.8 cm to be the bidealQ change in pain. It should be pointed out that just because the 10-cm VAS and the CAS are both measured in centimeter it does not necessarily mean that 2 cm in the 10-cm VAS is equivalent to 2 cm in the CAS. For example, the middle of the scale could have an expanded range or a compressed range of responses. Our study involved a different population enabling the results of these studies to be more generalizable.

Our results suggest that in order for a child to state that his or her pain is a blittle less,Q a mean decrease of 2.4 cm (95% CI, -3.15 to -1.72) on the CAS is required. This was defined as a clinically significant decrease in the child’s pain and is larger than that reported by Bulloch and Tenenbein where the mean decrease was 1.7 cm, with 95% CI of 1.1 to 2.2 cm.

In order for a child to state that the pain is much better, our results suggest that a mean decrease on the CAS of

5.4 cm is required. This was defined as the bidealQ change in pain and differs from Bulloch and Tenenbein who found a decrease in score of 3.8 cm (95% CI, 3.1-4.5 cm) to be ideal. It is likely that their point estimate is more accurate as they had 71 children who achieved the bmuch betterQ designation vs 29 in our study. What becomes evident in both studies is that a much larger change on the pain scale is needed (more than double) for children to feel that their pain

is much less, compared with what is required for their pain to be a little less.

These results can be used in the clinical setting when assessing the adequacy of therapeutic management and also by future investigators needing a quantitative standard to measure pain relief interventions in Pediatric populations.

Although this study was performed in an urban pediatric ED, the population consisted of mainly Caucasian and Hispanic children. It is recognized that sex, ethnicity, and age may play a role in pain reporting as may in previous experiences [12-14]. We did not inquire about any past painful episodes in the children’s lives. A much larger sample size would be required to explore the role each of these factors play. In addition, we only studied children with acute pain so our results are not applicable to children with chronic pain.

There are many important variables involved when assessing significant pain differences. Most published studies, including this one, have followed the same type of assumption set, which are serial pain assessments in the same patient. Whenever serial Pain measurements are used, the sampling is nonindependent because the pain values for a given patient are all linked with each other. Thus, the bsignificant pain differenceQ value only applies to serial pain measurements in the same patient and specifically excludes the comparison of pain between different individuals.

There were a total of 21 children who reported their pain to have increased from baseline. This may have been due to ineffective pain control measures at that time. Only one child judged their pain to be bmuch worseQ even though his/ her CAS score did not change.

Of the 126 pain comparisons made, 102 (81%) compar- isons were in children who initially described their pain intensity as moderate or severe. These children may have perceived more room on the scales to decrease their pain scales than to increase them [4]. This bceiling effectQ is intrinsic to all pain scales. However, an attempt was made to prevent this bias by not reminding children of their previous pain scores and by changing the location of the marker on the sliding scale upon each presentation to the child.

Only children aged 5 to 12 years were included because of the use of the Likert scale. In addition, children were enrolled regardless of the etiology of their pain. In 48% of our subjects, the pain was due to trauma with the remainder being a consequence of nontraumatic conditions. However, the etiology of pain (trauma vs nontrauma) in adults has not resulted in different minimum clinically significant differ- ences in VAS scores [12]. This has never been examined in children and requires additional study.

Conclusion

A decrease on the CAS pain scale of 2.4 cm constitutes a clinically significant change in pain for children aged 5 to 12 years with acute pain. The CAS is a valuable tool

in assessing responses to pain interventions when limited to serial pain scores in individual patients (and not pain scores between different groups). In addition, these results are specific to the CAS, which is not necessarily exactly equivalent to other pain scales that use the centi- meter measurement.

References

1. Joint Commission on Accreditation of Healthcare organizations. Comprehensive accreditation manual for hospitals. Oakbrook Terrace (IL)7 Joint Commission on Accreditation of Healthcare Organizations; 2001.
2. Drendel AL, Brouseau DC, Gorelick MH. Pain assessment for pediatric patients in the emergency department. Pediatrics 2006; 117(5):1511 - 8.
3. Zempsky WT, Cravero JP, Committee on Pediatric Emergency Medicine and section on Anesthesiology and pain medicine. Relief of pain and anxiety in pediatric patients in emergency medical systems. Pediatrics 2004;110:e33.
4. Todd KH, Funk KG, Funk JP, Bonacci R. Clinical significance of reported changes in pain severity. Ann Emerg Med 1996;27:485 - 9.
5. Bulloch B, Tenenbein M. Validation of two pain scales for use in the pediatric emergency department. Pediatrics 2002;110:e33.
6. Bulloch B, Tenenbein M. Assessment of clinically significant changes in acute pain in children. Acad Emerg Med 2002;9:199 - 202.
7. Eland JM, Anderson JE. The experience of pain in children. In: Jacox AK, editor. Pain: a sourcebook for nurses and other health professionals. Boston7 Little, Brown and Co.; 1977. p. 453 - 73.
8. Powell CV, Kelly A, Williams A. Determining the minimum clinically significant difference in visual analog pain scores for children. Ann Emerg Med 2001;37:28 - 31.
9. Wells GA, Tugwell P, Kragg GR. Minimum important difference between patients with rheumatoid arthritis: the patient’s perspective. J Rheumatol 1993;20:557 - 60.
10. Wright JM, Price SD, Watson WA. NSAID use and efficacy in the emergency department (letter). Anns Pharmacother 1994;28:1202.
11. Stahmer SA, Shofer FS, Marino A, et al. Do quantitative changes in pain intensity correlate with pain relief and satisfaction? Acad Emerg Med 1998;5:851 - 7.
12. Kelly AM. Does the clinically significant difference in visual analogue sale pain scores vary with gender, age or cause of pain? Acad Emerg Med 1998;5:1086 - 90.
13. Todd KN. Pain assessment and ethnicity. Ann Emerg Med 1996;27: 421 - 3.
14. Nevin K. Influence of sex on pain assessment and management. Ann Emerg Med 1996;27:424 - 6.