Metabolic components [body mass index (BMI), blood pressure, serum lipids] and physical activity may affect biological mechanisms of importance for breast cancer prognosis. A population-based survival study among 1,364 breast cancer cases within the Norwegian Counties Study during 1974-2005 was conducted. Pre-diagnostic measurements of BMI, blood pressure, serum lipids, and self-reported physical activity were assessed. Multivariable Cox proportional hazard models were used in analyses (SAS version 9.11). Among these breast cancer cases (age at diagnosis 27-79 years), 429 women died (8.2 mean follow-up years). Those with a BMI >or= 30 kg/m(2) had a 1.47 higher risk of dying during follow-up than women with a BMI of 18.5-25 kg/m(2) [hazard ratio (HR) = 1.47, 95% CI 1.08-1.99]. Women with BMI < 25 kg/m(2) and age of diagnosis >or=55 years had a 66% reduction in overall mortality if they regularly exercised before diagnosis compared with sedentary women (HR = 0.34, 95% CI 0.16-0.71). Women in the highest tertile of total cholesterol had a 29% increase in mortality compared to women in the lowest tertile (HR = 1.29, 95% CI 1.01-1.64). Additionally, women in the highest tertile of blood pressure had a 41% increase in mortality compared to women in the lowest tertile of blood pressure (HR = 1.41, 95% CI 1.09-1.83). Our study supports a relationship between mortality not only in relation to BMI, but also blood pressure, lipids, and physical activity among breast cancer patients. These factors may all be important targets for invention among breast cancer patients.