Pain in vulnerable populations unable to provide verbal report is challenging in terms of measurement and treatment. Clinicians strive to provide the best possible pain management for infants in the NICU, yet they are often hindered due to paucity of measures that are not only reliable and valid but also clinically useful. Clinical utility of measures is difficult to establish due to a lack of consistent definition of the construct, varied methods of determination, and the secondary importance afforded to this issue in relation to the establishment of reliability and utility. Without clinically useful pain measures, however, clinicians are unable and unlikely to assess the infant's pain or the effectiveness of pain-relieving interventions. Furthermore, even when the clinician is able to assess pain using a valid measure with a minimum of time, cost, and instruction, the clinical significance of any reduction in pain scores needs to be interpreted in terms of the infant and his/her care provider. The issue of defining the extent of change in pain scores that is clinically significant or important remains unclear. Clarity will involve assigning meaning to particular changes in pain scores for vulnerable infants across a broad array of situations and severities of pain. Although research on this topic in children and adults provides some guidance to this dilemma, only through innovative and creative methods will we be able to address these issues.