Comparison of Three Rating Scales for Measuring Subjective Phenomena in Clinical Research: I. Use of Experimentally Controlled Auditory Stimuli

An interest in measuring subjective phenomena such as pain, nausea, anxiety, etc. has led clinicians to develop three types of ratings: the visual analog scale (VAS); the verbal rating scale (VRS), and the numeric rating score (NRS). These ratings are regarded as global scales because they lack criteria to demarcate diverse dimensions or categories that comprise each scale. The purpose of this study was to evaluate validity and consistency of usage for these scales. Criterion for validity consisted of an experimentally controlled intensity for auditory stimuli. We conducted a prospective, experimentally controlled, clinimetric study at the Audiology Department at the Hospital of Puebla Autonomous University (in Puebla State, Mexico). Participants included 25 medical students, two psychology students, and three practicing physicians. Interventions consisted of pure 1,000 Hz tones in five different intensities applied for 3 sec with a 1-min interval between stimuli at three sessions for each observer. Main outcome measure was validity and consistency of usage for VAS, VRS, and NRS scales. Correlation coefficients between scale results and standard stimuli were 0.818 for VAS, 0.735 for NRS, and 0.796 for VRS. Mean weighted kappa indices for intraobserver agreement were 0.70, 0.59, and 0.65, respectively, for scales with five categories each. Mean weighted kappa indices for inter-observer variability were 0.61, 0.48, and 0.54 for VAS, NRS, and VRS again with five categories each. The three instruments appeared reasonably accurate, with VAS having highest scores. VRS appeared sufficiently consistent to be regarded as providing reliable scientific information.