© 1992 by Oxford University Press
Journal of the National Cancer Institute, Vol. 84, No. 11, 867-872,
June 3, 1992
© 1992 Oxford University Press
Case-Control Study of Factors Associated With Human T-Cell Leukemia Virus Type I Infection in Southern Miyazaki, Japan
Department of Epidemiology, Harvard School of Public Health, Harvard School of Public Health Boston, Mass
Second Department of Medicine, Miyazaki Medical School Kiyotake, Miyazaki, Japan
*Correspondence to: Sherri O. Stuver, Sc.D., Department of Epidemiology, Harvard School of Public Health, 677 Huntington Ave., Boston, MA 02115.
Background: An unusual age- and sex-specific distribution and a remarkably restricted geographic seroprevalence characterize human T-cell leukemia virus type I (HTLV-I) infection. Although the transmission routes of HTLV-I are known, these seroepidemiologic features cannot be fully explained. Purpose: This study was designed to identify potential characteristics associated with HTLV-I infection in a highly endemic Japanese community. Methods: We evaluated occupational, residential, dietary, and medical histories in a case-control study conducted in two neighboring villages in southeastern Miyazaki Prefecture. One hundred forty-four case subjects and 276 control subjects, frequency matched by age, sex, and village, were interviewed. Village-specific profiles of demographic determinants of HTLV-I seroprevalence were generated using multiple logistic regression. Results: Although a different pattern of factors was found for each village, occupations of the subjects and their fathers were associated with HTLV-I infection in both communitiesfarming in village A and fishing and farming in village B. For village A, there was more than a twofold association both with residence in the township for 55 years or more and with living in a particular area within the village. In addition, case subjects were more likely to have a mother who was deceased (odds ratio = 1·7; 95% confidence interval = 0·962·9). Conclusions: HTLV-I infection is characterized by a high degree of microepidemicity in this population, with seroprevalence related to both sociologic and geographic determinants. Moreover, as carriers' mothers themselves have a higher probability of being HTLV-I positive, an increased mortality among those infected with the virus is suggested. [J Natl Cancer Inst 84: 867872, 1992]