gms | German Medical Science

Background: To protect the privacy of clinical study or register participants, pseudonymization is used to store medical data without directly identifying information [1]. Pseudonymization describes the process of decoupling a patient's medical and identifying information while appending a patient-specific, non-speaking identification code to the medical data [2]. This enables researchers to link medical data in longitudinal research projects to the same patient. Pseudonymization is often performed by a trusted third party that receives the identifying information and returns a pseudonym [3].

To achieve this informational separation without forcing medical staff to operate two independent systems, we developed an extension for the electronic data capture (EDC) system REDCap [4] that integrates the open-source pseudonymization service Mainzelliste [5].

The extension was developed in the context of the Safety of the Living Kidney Donor – German National Register (SOLKID-GNR). This is a multi-centric register to explore the medical long-term effects of living kidney donors. When fully initiated, 38 hospitals throughout Germany will periodically interview living kidney donors regarding their physical and psychological well-being.

Methods: We utilized REDCap's External Modules framework to develop an extension that integrates Mainzelliste into REDCap [4]. The extension was built using HTML, CSS, JavaScript, and PHP. We paid attention to a high level of customizability, meaning that the extension should be easily adjustable to any Mainzelliste configuration directly within REDCap without knowledge of programming.

An instance of the pseudonymization service Mainzelliste is hosted by a trusted third party. The Institute of Medical Informatics hosts a REDCap instance for the medical data. Mainzelliste provides a REST interface for HTTP-based pseudonym generation [5]. In Mainzelliste, we configured the identifying data fields (i.e., first name, last name, etc.) and the authorized remote host (i.e., the IP address of the REDCap instance) for session, token, and pseudonym creation (cf. [5]).

Results: We developed and published an easily customizable Mainzelliste extension that natively integrates into REDCap. The backend is responsible for session and token creation. Once a token is generated that is eligible for pseudonym generation, this token is sent to the frontend. The frontend renders a form based on the configured identifying information fields. Once a user enters this information, the web browser sends a request to the Mainzelliste instance to request a pseudonym. Thereby, the identifying information is never sent to the backend. Afterwards, the extension redirects automatically to the respective record in REDCap.

The extension is openly available at REDCap's external module repository and GitLab [6].

Conclusion: REDCap is an EDC system that is currently cited in over 9.000 publications [7]. Even though it supports many features that are relevant for clinical studies and registers, native support for pseudonymization is currently not present. REDCap offers the opportunity to develop external modules to enhance its functionality. A repository helps to find modules from other developers. Nonetheless, an external module that integrates a pseudonymization service could not be found. Therefore, we developed an external module for REDCap that integrates Mainzelliste and made it publicly available.

Acknowledgment: Funded by a grant from BMBF for SOLKID-GNR (Grant no. 01GY1906).

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.