OBJECTIVES: There is currently no consensus on the appropriate method to estimate utilities for joint health conditions. We reviewed the literature to understand reasons for differences in conclusions drawn and to identify where further research is required. METHODS: We conducted a systematic literature search to identify studies that evaluated methods used to estimate mean utilities for comorbidities using mean values from cohorts with the corresponding single conditions. We extracted the preference-based utility measure used, the number and range of estimated utility values, the baseline used to value utility decrements, the statistics used to compare estimates, and the conclusions of the authors. RESULTS: Four of the six studies identified used EQ-5D data, one used SF-6D and one used HUI3. One presented the multiplicative method, one compared the additive and multiplicative methods, and four compared the additive, multiplicative, and minimum methods with the results obtained from linear models. The number of mean utility values estimated ranged from 32 to 760 and the range of actual mean values ranged from 0.465 to 0.607 for SF-6D, to -0.01 to 1 for HUI3.  Systematic errors were observed in the values estimated using all methods.  While the simple linear models produced the most accurate results these require validation. Of the other three, on average the multiplicative method estimated the most accurate values across the full range of actual utilities assessed. CONCLUSIONS: While additional research is required before a particular method can be advocated, based on the current evidence base we would recommend the multiplicative method is used if data are not available from cohorts with the comorbidities.