International comparison of subjective health evaluation: USA, UK and Japan

This paper tries to calculate quality of life (QOL) from subjective health evaluations in Japan following Cutler and Richardson (1997) and Groot (2000). It then extends the model in several ways. Firstly, while previous studies define the domain of QOL in an ad hoc manner, i.e. excluding 'excellent' or 'very poor' respondents, this paper suggests a more rigorous alternative measure. Secondly, heterogeneity among individuals that is inevitable in micro-data is accounted for in the estimation process. Thirdly, economic variables such as income or job status that are considered to affect subjective health status are also accounted for. The estimation results show the following: using the same model as previous work, similar tendencies are found, but coefficients are smaller for many symptoms and diseases. Economic variables help to clarify the effect of symptoms or diseases on subjective health evaluation. The QOL measures defined in this paper are smaller for most symptoms and diseases, and thus the measures in previous research are likely to overestimate of damage to QOL by symtoms and diseases and may be inappropriate.