Iwata K, Yamasaki T, Saito H, Ito S, Fukahori M, Nakazato T, Nagao S
Department of Preventive Medicine and Health Promotion, Nagasaki University School of Medicine.
Nihon Koshu Eisei Zasshi. 1994 Nov;41(11):1065-72.
Possible risk factors associated with mortality were studied in a community using data derived from annual mass health examinations for the aged mandated by law. A total of 1,804 adults (685 men and 1,119 women) aged 40 or older in A-town, located on Tsushima Island, Nagasaki Prefecture, Japan who had participated in annual health examinations at least once between 1984 and 1990, were followed for a mean period of 4.9 years. After adjustment for age using Cox proportional hazards models, in men liver dysfunction (aspartate aminotransferase > 40 U/l or alanine aminotransferase > 35 U/l), fasting blood glucose > or = 110 mg/dl and glucosuria, and in women serum creatinine > or = 1.2 mg/dl, fasting blood glucose > or = 110 mg/dl and proteinuria were found to be associated with a significantly increased risk of total mortality. In multivariate analysis using all independent variables that were significantly associated with mortality in age-adjusted bivariate analysis, in men liver dysfunction and hyperglycemia, and in women hypercreatininemia and hyperglycemia, were significant predictors of mortality. These independent variables remained significant or marginally significant predictors of total mortality even after excluding the effects of 3 pancreatic cancer cases with liver dysfunction or hyperglycemia or 12 deaths within the first year of follow-up, being associated with at least two-fold increased hazard rate ratios. From these results, it is recommended that persons with these risk factors be followed intensively and counseled by public health personnel to modify risk factors.
利用法律规定的老年人年度大规模健康检查数据,在一个社区中研究了与死亡率相关的潜在风险因素。对日本长崎县对马岛A镇1804名年龄在40岁及以上的成年人(685名男性和1119名女性)进行了研究,这些人在1984年至1990年期间至少参加过一次年度健康检查,平均随访4.9年。使用Cox比例风险模型对年龄进行调整后,发现男性肝功能障碍(天冬氨酸转氨酶>40 U/l或丙氨酸转氨酶>35 U/l)、空腹血糖≥110 mg/dl和糖尿,以及女性血清肌酐≥1.2 mg/dl、空腹血糖≥110 mg/dl和蛋白尿与总死亡率风险显著增加相关。在使用年龄调整后的双变量分析中与死亡率显著相关的所有自变量进行的多变量分析中,男性的肝功能障碍和高血糖,以及女性的高肌酐血症和高血糖是死亡率的显著预测因素。即使排除3例伴有肝功能障碍或高血糖的胰腺癌病例或随访第一年内的12例死亡的影响后,这些自变量仍然是总死亡率的显著或边缘显著预测因素,其危险率比值至少增加两倍。根据这些结果,建议对有这些风险因素的人进行密切随访,并由公共卫生人员提供咨询以改变风险因素。