The present study was performed to investigate the relationship between diabetes mellitus (DM) and primary hepatocellular carcinoma (HCC). Incident HCC cases were recruited in Kyushu, Japan. Ethnicity-, age-, gender-, residence-matched hospital controls and community controls were collected. Information on viral hepatitis B (HB5Ag) or viral hepatitis C infection, history of blood transfusion, past histories including DM, amount of drinking or smoking, and genotypes of alcohol metabolizing enzymes was collected. Associations between these items and HCC were analyzed multivariately by conditional logistic regression analysis. Two hundred and twenty two (177 males and 45 females) case-control sets were completed between July 1995 and June 2000. Since hospital controls turned out to be a biased one or those sampled from a DM-prone population, a multivariate analysis was performed for the HOC-community controls sets, and it yielded significantly elevated odds ratio (OR)s due to past histories of DM (2.522; 95% Confidence Interval (CI)=1 .267-5.020), blood transfusion (1.747; 1.136-2.689), and unit increment of alcohol consumption (1.358; 1.096-1.684) for males. The same analyses of the HOC-community controls sets for females, revealed an elevated but not statistically significant OR due to past histories of DM (4.195; 0.808-21.805). A multivariate analysis revealed that DM might be a risk factor for HOC.

A comparison of responses obtained in interviews using a structured questionnaire and those obtained using a self-administered questionnaire about parental history of liver diseases and personal smoking and drinking habits was made to evaluate whether the self-administered questionnaire is an effective means for collecting information on these variables in place of an interview. The subjects used in the present analyses were 46 male hepatocellular carcinoma (HOC) cases and 46 hospital controls who were interviewed between January, 1996 and January, 1997 for HOC cases and between January, 1996 and April, 1997 for hospital controls. Ages at start or stop of smoking and drinking were grouped into four categories as categorical variables. The cumulative number of cigarettes and cumulative alcohol consumption were also grouped into four categories. Generally speaking, the degree of agreement was substantially high for both categorical and continuous variables, particularly for parental history of liver diseases, and smoking (ICC=0.929 and 0.912 for cases and controls, respectively) and drinking (ICC=0.899 and 0.882, respectively) habits that are originally nominal categorical variables. On the other hand, the cumulative number of cigarettes (ICC=0.928 and 0.716, respectively) and the cumulative alcohol consumption (ICC=0.761 and 0.825, respectively) seemed to have a lower degree of agreement than other variables. For these variables, there was a tendency for the self-administered questionnaire to produce larger values compared with the interview. Therefore, the self-administered questionnaire seemed to be an effective means for collecting information on categorical variables, while the interview using a structured questionnaire seemed to be preferable to the self-administered questionnaire with regard to the lifetime exposure to cigarettes and alcohol.