OBJECTIVES: To empirically test whether and how framing of a risk attribute in a Discrete Choice Experiment (DCE) affects study results with respect to the relative importance of the attributes, trading behavior and potential uptake rates. METHODS: By means of ongoing data collection, two versions of a DCE- questionnaire containing nine D-efficiently designed choice tasks were distributed among a representative sample of the Dutch population aged 55-65years. The DCE consisted of four attributes related to the decision whether to participate in genetic screening for colorectal cancer (CRC). Three fixed attributes were; risk of being genetically predisposed, risk of developing CRC, and frequency of follow-up colonoscopies. The included risk attribute was framed positively as survival rate and negatively as mortality rate. Mixed logit models were conducted to estimate the relative importance of the attributes. Dominant decision behavior was determined and potential uptake rates were calculated. RESULTS: Overall, risk attribute framing significantly interacted with most of the attribute level estimates. Based on the positive frame, the frequency of follow-up colonoscopies was most important followed by survival rate, while based on the negative fame, mortality rate was most important. Twice as many respondents dominated on the survival attribute compared to the mortality attribute. Potential uptake rates were calculated for multiple hypothetical scenarios, in all cases they were lower based on the data of the negative frame. CONCLUSIONS: The use of a positive frame leads to significantly increased frequency of dominant choices. Negative framing of the risk attribute resulted in a different relative importance of the attributes and a lower willingness to participate in genetic screening for CRC compared to positive framing. These results call for greater attention and more research with regard to the impact of framing of risk attributes in DCEs aiming to elicit preferences within the health care or public health context.