Distributed data processing for public health surveillance

Lazarus, Ross; Yih, Katherine; Platt, Richard

The benefit of decreased risk of inadvertent PHI disclosure from our approach entails three principal disadvantages compared with routine, centralized collection of identifiable data. First, a clinical responder with access to the locally stored PHI data must be available to provide case level information when a cluster is detected. It would be technically straightforward to provide detailed information for relevant cases automatically when signals are detected. We deliberately did not implement this feature in the current system, since the participating health plans expressed a strong preference for direct involvement in this process.