Comparison of Two Pain Scales for the Assessment of Pain in the Ventilated Adult Patient

Abstract

The assessment of pain experienced by the patient requiring mechanical ventilation can be a challenge for critical care nurses. This study was conducted to assess the reliability and validity of two pain behavior assessment tools. Patients who met the inclusion criteria were enrolled from 13 critical care units and one long-term acute care unit within eight hospitals. A total of 200 patients were assessed over an 8-week period. The findings suggest that both tools have adequate reliability and validity. Use of a pain behavior tool is one strategy to improve assessment of pain experienced by the mechanically ventilated adult patient who is unable to provide a self-report.

A Validated Approach to Evaluating Psychometric Properties of Pain Assessment Tools for Use in Nonverbal Critically Ill Adults

“…46 The NVPS-I was tested in 123 ICU patients, 45,47 and the NVPS-R was tested in 264 ICU patients who were unable to self-report. 25,47 Cronbach α coefficients were lower (α ¼ 0.36 to 0.62) for the NVPS-I, 45,47 as compared with the NVPS-R (α ¼ 0.36 to 0.75). 25,47 Interrater agreement between ICU nurses was high (> 90%) for both versions of the tool evaluated by Kabes et al 47 but was lower (ICC 0.60 to 0.76) for the NVPS-R in Juarez et al 25 Discriminant validation was supported, with significant increases of 2 points in both versions of the scale during painful procedures (i.e., suctioning, turning, or repositioning).…”

“…25,47 Interrater agreement between ICU nurses was high (> 90%) for both versions of the tool evaluated by Kabes et al 47 but was lower (ICC 0.60 to 0.76) for the NVPS-R in Juarez et al 25 Discriminant validation was supported, with significant increases of 2 points in both versions of the scale during painful procedures (i.e., suctioning, turning, or repositioning). 25,47 There were several shortcomings of the study by Kabes et al: (1) the ICU patient sample was not well described (i.e., ICU primarily serving trauma and surgical patients but no more detail was given), (2) the observation time period for each assessment was not specified, (3) dependent data (i.e., data from the same patients over more than one assessment) were used in statistical analyses, (4) post hoc tests were not reported, and (5) the change in the respiratory item was not statistically supported. 47 A recent implementation study of the NVPS-I showed a significant increase in the number of pain assessments performed (2.2 vs 3.4 assessments per patient per day, p ¼ 0.02), and patients reported an overall decrease in their pain (8.5 vs 7.2, p ¼ 0.04) following implementation of the tool.…”

“…22 Cronbach α coefficients for internal consistency of BPS ranged from 0.63 to 0.72 in different studies. [23][24][25][26] Interrater reliability was moderate to high in most studies, with kappa coefficients ranging from 0.67 to 0.83 15,23,27 and an intraclass coefficient (ICC) of 0.95. 24 However, interrater agreement was lower in two studies, 25,26 with ICC from 0.46 to 0.73, and percentages of agreement ranged from 36 to 46% during pain assessments following nociceptive procedures (e.g., repositioning).…”

“…[23][24][25][26] Interrater reliability was moderate to high in most studies, with kappa coefficients ranging from 0.67 to 0.83 15,23,27 and an intraclass coefficient (ICC) of 0.95. 24 However, interrater agreement was lower in two studies, 25,26 with ICC from 0.46 to 0.73, and percentages of agreement ranged from 36 to 46% during pain assessments following nociceptive procedures (e.g., repositioning). Criterion validation was explored in one study, and a Spearman correlation coefficient of 0.67 (p < 0.001) between BPS scores and the patient's self-report of pain intensity (VRS [verbal rating scale]-4) during a painful procedure (turning) was obtained from 49 repeated observations within 30 subjects.…”

“…Discriminant validation was supported, with significant increases of 2 to 3 points in BPS scores following nociceptive procedures (e.g., turning, repositioning, endotracheal suctioning, peripheral venous cannulation), compared with BPS scores at rest or following non-nociceptive procedures (i.e., compression stocking applications, central venous catheter dressing changes, eye care, measuring body temperature). 15,[22][23][24][25][26] Most clinicians (86% of n ¼ 28 who used the BPS for the purpose of research) were satisfied by the ease of use of the tool, and all of them agreed that it took minimal time to complete. 15 Following implementation of the BPS tool, there were observed improvements in pain management, including an increase in the frequency of pain assessments and improved patient outcomes (i.e., shorter duration of mechanical ventilation and lower rate of nosocomial infections).…”