Objective: The study compared the preferences of patients and health-care professionals for the key attributes of a pharmacogenetic testing service to identify a patient's risk of developing a side effect (neutropenia) from the immunosuppressant, azathioprine. Methods: A discrete choice experiment was posted to a sample of patients (n=309) and healthcare professionals (HCPs) (n=410), as part of the TARGET study. Five attributes, with four levels each, described the service as follows: level of information given; predictive ability of the test; how the sample is collected; turnaround time for a result; who explains the test result. Data from each sample were first analyzed separately and responses were compared by 1) identifying the impact of the scale parameter, and 2) estimating marginal rates of substitution. Results: The final analysis included 159 patients and 138 HCPs (50% & 34% response rates). Estimated attribute coefficients from the patient and HCP sample differed in size, after taking into account the impact of the scale parameter. Patients and HCPs had similar preferences for predictive accuracy of the test and were willing to wait 2 days for a 1% improvement in test accuracy. Patients preferred to obtain more information and were willing to wait 19 days compared to 8 days for HCPs for providing higher levels of information. Conclusions: Patients demanded accurate and timely information from health-care professionals about why it was necessary to have a pharmacogenetic test and what the test results mean. In contrast, health-care professionals appear to focus more exclusively or entirely upon the predictive accuracy and waiting time for a test result.