C-reactive protein (CRP), an exquisitely sensitive systemic marker of inflammation, has emerged as an independent predictor of cardiovascular diseases (CVD). Because other chronic diseases are also associated with an inflammatory response, we sought to assess the association of high-sensitivity CRP (hsCRP) with total and cause-specific mortality in a large cohort of middle-aged men. METHODS: We measured hsCRP at baseline in 3620 middle-aged men, randomly drawn from 3 samples of the general population in the Augsburg area (Southern 0Germany) in 1984-85, 1989-90, and 1994-95. Outcome was defined as all deaths, fatal CVD, fatal coronary heart disease (CHD) including sudden cardiac deaths, and cancer deaths. RESULTS: During an average follow-up of 7.1 years, 408 deaths occurred (CVD 196, CHD 129, cancer 127). In multivariable Cox regression analysis, subjects with hsCRP >3 mg/L at baseline showed an almost 2-fold increased risk to die vs those with hsCRP <1 mg/L [hazard ratio (HR) 1.88, 95% CI 1.41-2.52]. HRs were 2.15 (95% CI 1.39-3.34) for fatal CVD, 1.74 (1.04-2.92) for fatal CHD, and 1.65 (1.01-2.68) for cancer mortality. In contrast, neither total nor HDL cholesterol significantly predicted all-cause or cancer mortality, and cholesterol had only modest effects on CVD mortality. CONCLUSIONS: Our results suggest that increased circulating hsCRP concentrations are associated with an increased risk of death from several widespread chronic diseases. Persistently increased hsCRP is a sensitive and valuable nonspecific indicator of an ongoing disease process that deserves serious and careful medical attention.