Biostatistics in the United States and Japan.

Abstract

This issue of ENVIRONMENTAL HEALTH PERSPECTIVES includes the proceedings of a conference sponsored by the Japan Society for the Promotion of Science and the U.S.-Japan Cooperative Cancer Research Program. The conference brought together American biostatisticians with Japanese mathematicians and epidemiologists to exchange ideas in statistical methods and approaches applied in cancer research in the United States and Japan. Interestingly, the conference helped to provide a formal introduction, perhaps for the first time, between the mathematical statistics and the medical (cancer) research communities inJapan, with the American contingent in effect serving as the catalyst. The field of statistics as an independent discipline has not developed in Japan as it has in the United States. In the U.S. there are over 190 universities offering graduates degrees in statistics, about 35 of which specialize in the field of biostatistics (1). In contrast there are no independent departments of statistics in any Japanese university, although mathematical statistics is taught under the mathematics curriculum in several of Japan's national and private universities. The primary reason for the absence of statistical units in Japan appears to lie in the vertical structural organization of Japanese universities (and other institutions as well). Interaction between different departments is minimal, reflecting a general societal pattern in which individuals tend to associate primarily with those in their own group rather than with individuals, even of similar attributes, in other groups (2). Since biostatistics is a discipline traditionally formed by the merger of different groups mathematics and statistics on the one hand, biology and medicine on the other its de-