Research Roadmap

Research data is the foundation of scientific research. Proper management of this data is important to prevent loss, unintentional changes, or manipulation. Legislation and an increasing emphasis on the reproducibility and integrity of research require transparent working procedures. Funding bodies also increasingly set specific requirements for data management.

The department Research Data Management (RDM) provides comprehensive information, advice, and support on this matter. They have developed procedures and templates and a SOP Research Data Management, which describe both legal requirements and Amsterdam UMC standards for research data management. These resources also offer guidelines and practical tips for handling research data — from collection and storage to processing and archiving. A few specific topics are highlighted below:

Data management plan

Before starting your research, you are required to prepare a detailed Data Management Plan (DMP), which demonstrates how your research complies with the requirements set out in the SOP Research Data Management. The DMP should specify the types of research data to be collected, how that data will be stored and managed during the study, and what will happen to the data after the study is completed. A DMP is an important tool for safeguarding data integrity. The Amsterdam UMC template is available here.

For questions about research data management or drafting a DMP, consult the website of RDM or contact the RDM helpdesk via the RDM Service Portal

Preparing for data collection

When collecting personal data for research subject to the WMO, the following (legal) principles apply:

• Never collect more personal data than needed for your study;
• Obtain written consent from participants for collecting their personal data;
• Never use or share personal data for purposes other than those consented to by particpants;
• Ensure that collected data is limited to the research data as specified in the research protocol (data minimisation).

Data can be collected in different ways:

• Reuse existing data from research databases, registries or the Electronic Patient Dossier (EPD);
• Create a new dataset by manually entering the data from participants in a database;
• Generate data by using a machine or piece of equipment, such as laboratory experiments, DNA sequencing or radiology images.

Reuse of healthcare data

Reusing existing data avoids unnecessary duplication and minimizes errors. A great deal of structured healthcare data is available for research via the Research Data Platform (RDP). However, legal provisions restrict the reuse of such data for research without meeting specific conditions, including respecting privacy and patient autonomy. The conditions that must be met are described in the SOP on the Reuse of care data for the purpose of research. Consult the website of RDM for more information.

Creating a new dataset

Amsterdam UMC offers various applications and tools for collecting and storing new research data. The website of RDM provides detailed information on using the right software for building databases (eCRF), setting up questionnaires or randomisation programs, creating data dictionaries, implementing validation controls, and more. RDM also offers courses in research data management and the application Castor EDC.

All data generated during research and archived after the study is completed must be safely stored while remaining accessible for the research team.

If existing healthcare data cannot be used, new data is often collected by recruiting participants through the clinic (inpatient or outpatient) or through advertisements, internet sites, leaflets, or other tools.

When collecting data for research, you will work with source data and a Case Report Form (CRF), either on paper or electronically (database). The DMP documents how data will be collected, stored, and validated. The statistical analysis plan documents the analytical methods to be used. More information on analytic databases is available on a dedicated page: Implementing statistical analyses.

Audit trail and the (e)CRF

A CRF (paper or electronic) is used to collect all research data for each participant. The guideline for Good Clinical Practice (GCP) requires that ‘Any change or correction to a CRF must be dated, initialled and (if necessary) explained, and it must not render the original text unreadable (i.e. an audit trail must be maintained). This requirement covers both manual and electronic changes or corrections’ (GCP Section 4.9.3).

Note: SPSS and Excel do not have an audit trail, and are therefore not suitable as CRFs/eCRFs.

Amsterdam UMC offers several options for electronic CRFs (eCRF), including access to Castor EDC (free of charge). Castor EDC can be used to perform randomisation, send electronic questionnaires to participants, give access to monitors of a study, export data to Excel/CSV/SPSS, and more.

Validation procedures

To ensure the quality of the dataset, data entry must be standardized and transparent. It is advisable to establish validation procedures to inspect data quality during entry. Consult the website of RDM for more information.

Randomisation

Random assignment of participants to a treatment arm in clinical trials (i.e. randomisation) should have minimal involvement of the researcher; computer-driven assignment techniques are preferred.

Castor EDC offers several randomisation methods that can be integrated into the eCRF. For more information, contact the RDM helpdesk via the RDM Service Portal.

Online Questionnaires

If you use online questionnaires outside Castor EDC, an agreement with the provider of the application/tool is required. Consult the website of RDM for more information, or contact the RDM helpdesk via the RDM Service Portal.