OBJECTIVE: The aim of our study was to evaluate the efficacy of the addition of lifestyle related variables to the existing Framingham risk score (FRS) for predicting cardiovascular death in patients with pre-existing cardiovascular disease (CVD). DESIGN: Survey analysis. METHODS: Participants were recruited from the Health Survey for England and Scottish Health Survey, both representative population-based studies sampling adults in each country. 1372 patients with CVD (age 66.4±10.9yrs, 54.7% male) were included. Parameter estimates from Cox proportional hazards models were used to derive a new global risk score (termed the Health Survey Risk Score; HSRS) to predict CVD death. Receiver operating characteristic curves were compared to the FRS algorithm. The addition of each risk factor to the conventional FRS model examined the discriminative power of alternative lifestyle-related variables. RESULTS: Cholesterol/HDL ratio, diabetes, smoking, C-reactive protein and moderate-vigorous physical activity (MVPA) were all independently associated with CVD death in males. In females, only age, diabetes, smoking, and BMI were significant independent predictors. The FRS risk factor model was predictive of CVD death (AUC=0.659; 95% CI 0.609-0.708 in males; AUC=0.689; 95% CI 0.621-0.756 in females). However, there was an additional benefit of including other lifestyle-related variables for the prediction of CVD death in CVD patients (AUC=0.751, 95% CI 0.710-0.792 in males; AUC=0.748, 95% CI 0.687-0.810 in females). CONCLUSIONS: The additional inclusion of lifestyle-related risk factors (body mass index, MVPA and psychological distress) and inflammatory marker (CRP) improved risk prediction of cardiovascular death in patients with pre-existing CVD, especially in males.