You can measure outcomes without specifics of who was impacted. The doctor actually as a bulwark in this case and only provides the data specific to the devices as a report. I.e. we performed 175 hip replacement surgeries using x device. During that time there was a median issue rate of x percent. This is within x range of standard. This would have the added benefit of evaluating the medical professionals against similar circumstances especially if you factors in removing outliers that have a higher risk profile.