1. The intensity of breathlessness during exercise was measured in ten normal subjects using a visual analogue scale (VAS) and a Borg scale to compare the use of the scales and their repeatability, both within the duration of a period of exercise and between tests. For each scale, subjects performed two exercise tests separated by a period of 2–6 weeks. Each exercise test consisted of two cycles of progressively increasing and decreasing workload.
2. All subjects felt confidently able to use both scales to quantify their feelings of breathlessness exclusively of other sensation. Equal preference was expressed for use of a particular scale.
3. With both scales there was a large intersubject variation in the relationship between dyspnoea score and minute ventilation (VE) (P < 0.01), and in the range of the scale used.
4. There was a good correlation between the VAS and Borg scores at each level of VE (r2 = 0.71), but the VAS score was used over a wider range than the Borg score.
5. The relationship between VE and the dyspnoea score measured by the two techniques was predominantly linear. The mean r2 for VAS score/VE was 0.68 (sd 0.19) and for Borg score/VE the mean r2 was 0.75 (sd 0.13).
6. The relationships VAS score/VE and Borg score/VE were unaffected by the direction in which the workload was varied (P > 0.05).
7. VE, measured at each work rate, did not differ between the two cycles (P > 0.05) or between the 2 days (P > 0.05).
8. With both scales, the slope of the VE-breathlessness relationship was slightly higher during the second half of the exercise compared with the first (0.05 < P > 0.01).
9. The scores with both scales were lower in the second test compared with the first (P < 0.01): Borg 16% lower, VAS 27% lower.
10. Measurements of dyspnoea made with the Borg scale appeared to have greater stability than VAS measurements and to correlate with VE a little better.