A comparison of three rating scales for measuring subjective phenomena in clinical research. II. Use of experimentally controlled visual stimuli.

BACKGROUND

In a previous study of three types of global scales we found that verbal rating scales were particularly reliable for rating auditory stimuli. We now wanted to check the performance of the scales for rating experimentally controlled visual stimuli.

METHODS

We used a prospective, experimentally controlled, clinimetric study, which was conducted at the Department of Psychiatry of the Autonomous University of Puebla Medical School in the state capital city of Puebla, Mexico. A total of 20 fifth-year medical students participated in the study. Visual stimuli consisted of 15 cards with five different intensities on the gray-to-black scale, administered randomly in three sessions to each subject. With regard to main outcome measurement, validity and consistency indices were determined for visual analog scale (VAS), numerical rating score (NRS), and verbal rating scale (VRS) to rate visual stimuli.

RESULTS

For validity, correlation coefficients between scales and reference standard were high, especially in VRS (r=0.902). For consistency, VRS had highest kappa value (k(w)=0.71) for interobserver variability.

CONCLUSIONS

Three instruments could be hierarchically ranked for their indices of validity and consistency. Being more consistent than VAS and NRS, VRS merits more frequent usage in clinical research.