Evaluating the Reliability, Validity, and Clinical Utility of Two Measures of Neonatal Pain

Abstract

Traditionally, neonates were thought to lack the capacity to perceive pain. Research has since refuted this notion, and there is now an urgent need for improved pain assessment practices, particularly in the neonatal intensive care unit (NICU). While there is an abundance of published neonatal pain assessment measures, few have undergone extensive psychometric testing, and there is a dearth of studies formally assessing their clinical utility. This study aimed to investigate the internal consistency, inter-rater reliability, concurrent and construct validity, and the clinical utility of the COVERS scale and the modified Pain Assessment Tool (PAT) within the NICU of Auckland City Hospital. It also aimed to establish the scales’ reliability and validity when used among extremely preterm, preterm, and term neonates. Eighty infants undergoing a routine blood-collection procedure were assessed at baseline and during heel lance by two, independent nurses using the two pain measures. Both scales had satisfactory internal consistency, and high agreement between the two raters. Total scores on each scale correlated highly, and scores correlated with pain ratings provided by mothers and the researcher, which supports both scales’ concurrent validity. Support for the construct validity of both scales was gained by results showing that the scales could discriminate effectively between pain and non-pain stimuli. Interestingly, neither scale had acceptable inter-rater reliability or concurrent validity for term infants at baseline, and the concurrent validity of both scales among extremely preterm infants during heel lance was moderate. Relative to the modified PAT, the COVERS scale also demonstrated weaker internal consistency among extremely preterm infants, and weaker inter-rater reliability among preterm infants. Following the study with infants, 25 nurse participants completed a questionnaire about their perceptions of the clinical utility of each scale. Both scales received positive evaluations from nurses. However, the modified PAT was rated as more appropriate to the range of infants admitted to the NICU, and was the preferred tool in this study. This research has implications for the clinical assessment of neonatal pain. The results suggest that a one-size-fits-all approach to neonatal pain assessment may not be feasible. Results also show that both the COVERS scale and the modified PAT represent viable means of assessing neonatal pain in the intensive care unit; however, the modified PAT is more suited to the NICU of Auckland City Hospital.