OBJECTIVES: Disease-burden measured by preference-based Health Related Quality of Life (HRQoL) scores is dependent on the HRQoL instrument used. We test a simple mapping algorithm to make disease-burden estimates across several HRQoL scoring systems comparable. METHODS: We used National Health Measurement Study (NHMS) data, a national sample of 3844 adults. We measured the burden associated with 31 health conditions, as captured by 8 HRQoL instruments (EQ-5D with UK and US scoring, HALex, HUI Mark 2 and Mark 3, QWB-SA, SF-6D[12 and 36]) by regressing HRQoL scores onto each condition, adjusting for age, sex, education, home ownership, and race. The regression coefficient for the health condition is the “disease-burden”. We then regressed disease-burden measured by each HRQoL measure for the 31 conditions onto those measured by the EQ-5D US. We test the resulting mapping algorithms using 2000 Medical Expenditures Panel Survey (MEPS) data (EQ-5D UK and SF-6D(12) to EQ-5D US) and US Valuation of the EuroQol EQ-5D Health States (USVEQ) data (HUI Mark2 and HUI Mark3 to EQ-5D US). RESULTS: Disease-burdens, i.e., HRQoL decrements, for the 31 conditions as measured by the EQ-5D US ranged from -0.002 to -0.195 on this scale. Modeling disease-burden from other systems to the EQ-5D US exhibited excellent fit (R2 all > .85). When tested in MEPS, the predicted disease-burdens in the EQ-5D US were slightly biased (0.0012 from SF-6D(12), 0.0083 from EQ-5D UK). Tested in USVEQ data, the predicted disease-burdens for the EQ-5D US were more biased (0.0125 from HUI Mark 2, 0.0144 from HUI Mark 3). The standard deviation of differences between predicted and actual disease-burdens was around 0.01 for most measures. CONCLUSION: This method produces relatively unbiased estimates for disease-burden across HRQoL scoring systems. Current error variance associated with this method limits its usability. Future work should examine non-linear cross-walks between HRQoL scores.