Patient data are fragmented across clinical and claim repositories with no single unique identifier. The demand for collaborations that use patient data is growing and with it the need to integrate data in a privacy-preserving fashion. Factors limiting the exchange of identified health data include a lack of a business associate relationship between collaborators, legislation, business processes, and sociopolitical expectations. Consequently, the use of privacy-preserving record linkage (PPRL) methods is growing. Evidence-based approaches for optimizing PPRL accuracy are needed.
