| In this dissertation, I study the mechanism through which surgeons accumulate peer reputation by signaling quality and its implications for their pace of technology adoption. Peer physicians make attributions about surgeon quality by disproportionately weighting an outcome's signal by its likelihood (nonlinear weighting). This attribution behavior creates an incentive for surgeons to actively manage case-mix. Empirical analysis supports the theoretical predictions that succeeding builds volume faster on difficult cases than on easy cases and failing depletes it more rapidly on easy cases than on difficult cases. A counterintuitive prediction also shown empirically is that case-mix decreases with increasing peer reputation. The final piece of theoretical analysis demonstrates, also counterintuitively, that high reputation surgeons will adopt more quickly than low reputation surgeons. In the empirical analysis, I find strong supportive evidence that surgeons with low severity case-mix adopt the new technology more rapidly. The policy implications include the need for physician groups to institute swift guidelines regarding adoption of new technologies and reconsideration of the information given in report cards. Furthermore, organizations should account for surgeons' private incentives in managing case allocation. |
