Public preferences for robotic surgery across two cancer scenarios: a latent class discrete choice study.

Public preferences for robotic surgery may reflect different responses to the technology, physician recommendation, cost, and clinical context. This study characterized latent preference profiles across colorectal and thyroid cancer scenarios using a joint model with shared class membership and scenario-specific utilities. A discrete choice experiment was conducted among 1,012 Korean adults. Colorectal cancer choices compared robotic and laparoscopic surgery, whereas thyroid cancer choices included robotic, endoscopic, and open surgery. Attributes included out-of-pocket cost, physician recommendation, and family or acquaintance influence; the thyroid scenario incorporated a scar-sparing profile. A joint latent class logit model shared class membership across scenarios while allowing scenario-specific utilities. Two- to four-class models were evaluated using multistart estimation, and a one-step model examined characteristics associated with class membership. The three-class model characterized context-sensitive (60.6%), technology-positive (29.1%), and conventional-surgery-oriented (10.3%) profiles (mean maximum posterior probability, 0.948). The technology-positive class favored robotic surgery in both scenarios but also responded to physician recommendation. The context-sensitive class had positive utility for the thyroid scar-sparing profile but negative incremental utility for robotic over endoscopic surgery. Higher technology acceptance (OR, 3.17; 95% CI, 2.49-4.03) and previous cancer experience (OR, 2.31; 95% CI, 1.20-4.45) were associated with technology-positive membership. Stated preferences among general adults for robotic surgery reflected distinct valuation profiles rather than a single pattern of acceptance. Apparent demand for robotic surgery may arise through different combinations of technology preference, physician recommendation, cost, and scenario-specific considerations. These demand-side preference patterns provide testable hypotheses about how public preferences may interact with provider adoption in the diffusion of high-cost surgical technologies.