We have previously reported increased mortality associated with the DXA-derived BCP, “barrel” (Z_%trunk_fat >0 and Z_limb_fat < 0) and “soft” barrel (“barrel” and Z_limb_lean < 0). The age (>20 years) and gender stratified study population (N = 324) from Malmö, Sweden, had LUNAR DPX total body scans at baseline. We define: 1. Z_%trunk_fat -% total fat in trunk, 2. Z_limb_fat - height corrected limb fat (kg/m2), and 3. Z_limb_lean -height corrected limb lean tissue (an index of skeletal muscle mass). Subjects were assigned a 1 if concurrent for the previously defined prediction of the BCP otherwise 0, to yield a predictive value for each BCP. BCP predictions were compared by paired T tests. We also examined using as a predictor, the projection of the Z-score coordinates onto the respective octant, to give a continuous BCP “risk score”. Analysis was repeated for cardiovascular mortality. Total mortality at 10 years was 20.4% (66/324). Mortality did not differ for the three single Z scores but was significantly higher for the barrel profile (27.8% vs. 17.2%, p = .03, and for the soft barrel profile (31.6% vs. 18.0 %, p= .02). Predictive value for soft barrel was higher than for barrel (t = 3.5, p < .001) and higher yet for the BCP risk scores (t = 8.1, p< 0.001). Also soft barrel profile predicted cardiovascular better than total mortality (p = .009). These results strengthen the association of barrel BCP and 10-year cardiovascular mortality in this population.