The difficulty of recruiting patients is a well-known issue in clinical trials which inhibits or sometimes precludes them in practice. We interpret this issue as a non-standard version of exploration-exploitation tradeoff: here, the clinical trial would like to explore as uniformly as possible, whereas each patient prefers ``exploitation", i.e. treatments that seem best. We study how to incentivize participation by leveraging information asymmetry between the trial and the patients. We measure statistical performance via worst-case estimation error under adversarially generated outcomes, a standard objective for clinical trials. We obtain a near-optimal solution in terms of this objective. Namely, we provide an incentive-compatible mechanism with a particular guarantee, and a nearly matching impossibility result for any incentive-compatible mechanism. Our results extend to agents with heterogeneous public and private types.

翻译：患者招募困难是临床试验中众所周知的问题，这一问题在实践中会阻碍甚至有时完全阻止试验的进行。我们将此问题解释为一种非标准形式的探索-利用权衡：在此情境下，临床试验希望尽可能均匀地探索，而每位患者则更偏好"利用"，即选择看似最佳的治疗方案。我们研究如何通过利用试验与患者之间的信息不对称来激励参与。我们以对抗性生成结果下的最坏情况估计误差来衡量统计性能，这是临床试验的标准目标。我们在此目标下获得了接近最优的解决方案。具体而言，我们提出了一种具有特定保证的激励相容机制，并证明了任何激励相容机制几乎匹配的不可能性结果。我们的结论可推广至具有异质性公开和私有类型的参与者。