OBJECTIVES A recent double-blind, placebo-controlled clinical study with lubiprostone in opioid-induced constipation (OIC), OBD-1033, included the EQ-5D generic quality-of-life instrument, and the PAC-QOL, a constipation-specific disease measure. This study calculated utility values for patients with OIC using the direct EQ-5D responses, and compared the resulting utilities to those calculated from a published mapping formula between the PAC-QOL and EQ-5D that was derived in chronic idiopathic constipation. METHODS EQ-5D responses from OBD-1033 were converted to utilities using the EQ-5D UK value set. These were compared with utilities generated with the published mapping algorithm. Following this step, an attempt was made to map the PAC-QOL to the EQ-5D in OIC. The root mean squared error (RMSE), adjusted R, and predicted/observed plots were used to assess the quality of mappings. RESULTS Patients in OBD-1033 had low utility values at baseline: mean=0.45 (Standard Deviation 0.33, n=439). Using the published algorithm, the predicted mean utility was much higher: 0.74. This led to a high RMSE (0.43), indicating a poor fit to the data. Replicating the mapping using OBD-1033 PAC-QOL and EQ-5D data showed the PAC-QOL, although correlated with the EQ-5D, had a poor predictive value (RMSE=0.31; R<0.10). High utilities were underestimated and low utilities overestimated. CONCLUSIONS Mapping algorithms are a vital tool for generating utility values when none are available. However, the relationship derived between instruments should be assessed cautiously. Mappings with the same instruments may not be reliable if crossing disease areas – even if the symptoms experienced by patients appear similar. Data show patients in OBD-1033 entered the study with poorer health status than those in the chronic constipation mapping (utility of 0.45 vs 0.81), likely due to comorbid conditions (the reason for opioid prescribing). This led to a different relationship between the PAC-QOL and EQ-5D, compared to the previous estimate.