Measuring pain intensity in older patients: a comparison of five scales

Abstract

Abstract Background and aims Pain is common in older individuals. In order to understand and treat pain in this group, reliable and valid measures are needed. This study aimed to evaluate: (1) the validity, utility, incorrect response rates and preference rates of 5 pain rating scales in older individuals; and (2) the associations between age, education level, and cognitive function and both (a) incorrect response and (b) preference rates. Methods Two hundred and one orthopedic clinic outpatients ≥ 65 years old were asked to rate their current pain, and least, average, and worst pain intensity in the past week using 5 scales: Verbal Numerical Rating Scale (VNRS), Faces Pain Scale - Revised (FPS-R), Verbal Rating Scale (VRS), Numerical Rating Scale (NRS), and Visual Analogue Scale (VAS). Participants were also asked to indicate scale preference. We computed the associations between each measure and a factor score representing the shared variance among the scales, the incorrect response and scale preference rates, and the associations between incorrect response and preference rates and age, education level, and cognitive function. The incorrect responses included being unable to respond, providing more than one response, responses outside a range, providing range answers rather than fixed answers, and responses indicating ‘least > average,’ ‘least > worst,’ and ‘average > worst’. Results The findings support validity of all 5 scales in older individuals who are able to use all measures. The VNRS had the lowest (2%) and the VAS had the highest (6%) incorrect response rates. The NRS was the most (35%) and the VAS was the least (5%) preferred. Age was associated with the incorrect response rates of the VRS and VAS, such that older individuals were less likely to use these scales correctly. Education level was associated with the incorrect response rates of the FPS-R, NRS and VAS, such that those with less education were less likely to use these measures correctly. Cognitive function was not significantly associated with incorrect response rates. Age, education level and cognitive function were not significantly associated with scale preference. Conclusions Although all five scales are valid, the VNRS evidences the best overall utility in this sample of older individuals with pain. The NRS or FPS-R would be fine alternatives if it is not practical or feasible to use the VNRS.