Generic health status has been recommended to be measured separately from disease-specific health status, because they can yield complementary information. In particular, generic health status can provide comprehensive health ratings across various disorders. However, the weakness with generic measures is that they may be less responsive to clinical changes than disease-specific ones. Therefore, when using generic health status as an endpoint in clinical trials, the instrument to be used is a problem with respect to responsiveness. In the present study, we investigated and compared the responsiveness of health status measures during asthma treatment using three different generic instruments: the Medical Outcomes Study Short Form 36-items Health Survey (SF-36), the Nottingham Health Profile (NHP) and the EQ5D (EuroQoL), as well as one disease-specific instrument, the Asthma Quality of Life Questionnaire (AQLQ). Fifty-four new patients with asthma who consulted our clinic were recruited. The health status measurements were performed on the initial visit, and at 3 and 6 months. All subscales of the SF-36 showed a significant improvement during the first 6 months. Each dimension of the EQ5D showed stronger ceiling effects than the SF-36. With respect to the responsiveness indices, the SF-36 was regarded as more responsive than the NHP or EQ5D utility. The changes in the SF-36 had a weak to moderate correlation with the changes in the AQLQ. In conclusion, the SF-36 had a higher responsiveness for asthma as a generic measure than the NHP or EQ5D, and evaluated different aspects from the AQLQ. The SF-36 can be used effectively in asthma clinical trials.