In this issue of JNN, Poulsen et al. present findings from a validation study done to develop a pain assessment scale (PAS) that could be used in the clinical setting for patients with disorders of consciousness (DOC) after an acquired brain injury. For this section of Research Bytes, we'll break down what was done in the study, review findings, and discuss what it means for nursing practice.

Significance

Assessing pain in patients with DOC after a brain injury can be extremely difficult. Typical pain scales, such as visual analog or Likert type scales, are often impossible for these patients to use when reporting pain. Although tools have been developed for assessing pain in other populations, such as neonates or patients with dementia, these tools may not be accurate for patients with acquired brain injury, who may exhibit different reactions to pain. Physiological indices of pain have been identified in patients with brain injury but have not cumulatively been linked together into an assessment tool for use in clinical practice. Therefore, the purpose of this study was to develop a potential tool for assessing pain among patients with acquired brain injury with DOC in the clinical setting.

Study Purpose

The aim of this study was to examine interrater agreement and sensitivity to change over time of a PAS for severely brain-injured patients with DOC.

Methods

Study investigators created a list of 27 indicators of pain or nociception exhibited among brain-injured patients based on findings from the literature, clinical experiences, and expertise. Indicators were categorized as physiological/autonomic, body language, verbal communication, and behavior and rated as present or absent. Adult patients admitted to neurointensive care or rehabilitation units with acute brain injury, who could not express themselves verbally, were included. Three nurses rated patients before and 5 minutes after repositioning and before and 30 minutes after analgesics.

Results

Data were collected on 26 patients, resulting in 44 ratings. Interrater reliability among nurses rating the patients was calculated using Cohen's kappa statistic, which evaluates degree of agreement among raters when controlling for chance. Kappa scores range from 0 to 1, with 1 indicating a perfect agreement. A kappa score of 0.6 is acceptable, with higher scores (0.75 and above) indicative of high reliability among raters. Kappa values were high (0.8 or above) for 13 items on the scale, meaning that values assigned by nurses for these items were almost always the same. Kappa scores were moderate (0.6-0.8) for eight items on the scale, indicating some degree of agreement among nurses, and three items on the scale had kappa values less than 0.6, reflecting poor agreement. Wilcoxon signed rank tests were used to examine how sensitive ratings were to change before and after repositioning and administration of analgesics. Analyses indicate that the scale was sensitive to change before and after repositioning but not before or after administration of analgesics.

Clinical Implications

This study presents preliminary findings from PAS developed to specifically measure pain in patients with DOC from acquired brain injury in the clinical setting. Initial findings suggest good agreement for most items on the scale and the possibility that the tool is sensitive to change; however, additional psychometric testing is needed to further refine the tool before routine use in clinical practice.