For clinical record management, patients typically assigned an id within a medical center or a clinical study (often along with a national id). A bottleneck in leveraging patient information for large-scale data analyzes is combining data across different clinical studies. Given the sensitive nature of patient data, it is hard to know which patients are the same in different studies or data sources.
It seems there are some efforts but nothing standard yet (sth like a universal id as there is for researchers, i.e., ORCID ). These include some software typically used within the data of the same healthcare provider or some initiatives that are being considered such as
- Managing Patient Identity Across Data Sources
- Implementing a hash-based privacy-preserving record linkage tool in the OneFlorida clinical research network
- Patient ID now
It is somewhat surprising that an additional one-way hash (combining various sensitive information from patient such as name, birthday etc.) is not stored in clinical/health records to facilitate linking of same individuals. Are you aware of any reliable publicly available / web-based resource that generates/records/validates such (non-identifiable) patient ids (either nation-wide or globally) toward identification of patients in different clinical studies (e.g., an ORCID but for patients without associating any identifiable information publicly)? Or would this be impossible to achieve due to current regulations in different countries?
