Rationale, aims and objectives: Recent analyses from the Florida Dental Care Study found that response to a hypothetical scenario at baseline strongly predicted: (a) tooth loss during follow-up; and (b) subsequent receipt of either a dental extraction or Root Canal Therapy (RCT). The scenario ('CHOICE') required choosing either to: (1) extract the tooth before even knowing the cost of treatments; (2) extract, but after knowing the cost of all treatments; or (3) have RCT despite knowing costs. Objective: The purpose of this study was to identify factors associated with CHOICE and quantify their effects. Methods: As part of the baseline phase of the study, 873 subjects with at least one tooth and who were 45 years or older participated for an interview and dental examination. A multinomial multivariable regression of CHOICE quantified effects due to hypothesized predictors. Results: CHOICE was strongly associated with race (African-Americans were significantly less likely to choose RCT). Results from the multivariable regression suggest that the race effect could be explained by racial differences in patient preference, treatment acceptability and ability to afford treatment. Conclusions: There were substantial racial differences in treatment preference even in this hypothetical scenario where racial differences in patient-provider interaction and clinical factors were not relevant. Certain predisposing and enabling variables explained these racial differences in treatment preference. © 2006 Blackwell Publishing Ltd.