Data Management Plan

Data Management Plan

A well written Data Management plan (DMP) should provide a quick reference to the most important aspects of study conduct. Created at the start of a study, the DMP forces complete study planning, playing a key role in ensuring efficiency and data integrity. Most companies have a DMP template as an appendix to an SOP, including generic text.

Electronic e-healthcare Data management plan

Under pressures of study set-up timelines, it is tempting for the DMP template to be updated with the minimum amount of information. However, any time saved at this stage will cause inefficiencies further down the line as the Data Manager trawls through supporting documents to find an answer.

The level of detail provided within the DMP varies between Data Managers and striking the right balance comes with experience.

Developed with the aid of the study protocol, CRF/ data collection forms, and SOPs/working instructions: key factors should be considered by both the author and reviewer.

  • Protocol version – The protocol should be finalised or near-final prior to the start of the DMP. Drafting the DMP too early poses the risk of irrelevant/redundant strategies being put in place, inaccurate information being reported, resulting in hours of re-work.
  • SOPS/Standards – The applicable SOPS and versions, as well as standard to be followed, should  be listed, i.e CRO’s, Sponsor’s, or a mix of both  SOPs to be followed, omitting these details could result in an incorrect process being followed.
  • Study timelines – More often than not a delay in the initial study set-up phases can threaten overall study timelines. The DM needs to have strategies in place to mitigate delays that could jeopardise the important first patient first visit date.  If a staggered EDC deployment approach is needed, the DMP should include the strategy, i.e. the number of deployments planned, visits/forms/ automatic edit checks to be included in each deployment.
  • Type and phase of study – Ensuring the data cleaning strategy for the study is documented. Considering that a phase III study would be different from that of a retrospective observational study, the strategy should be phase specific. This will avoid unrealistic/unnecessary expectations and ensure efforts are focused on key areas.
  • External vendor data – As well as referencing the Data Transfer specification developed for each vendor, the frequency of transfers, key contacts, and agreed process for receipt of unblinded data should be included.
  • Paper, electronic or hybrid study – Ensuring that all processes are looked at in view of the complete data collection strategy, i.e. if data is to be entered from paper questionnaires into the EDC system is a cross-check needed between the eCRF and questionnaires?
  • EDC/database to be used for data collection – Key features of the EDC/Database base should be included: Languages, user roles, version, etc. Including the version will help explain the lack of a certain functionality if a newer version is released and not adopted.
  • Data entry process – The complete data entry process should be included, from receipt of a paper document to data entry. Details on how will the data be received, how soon the data will be entered, type of data entry ( single/double), details of data entry QC to be performed, acceptable error rate. The transmittal form, Data entry report and other documents to be used for the process should be included in the DMP appendix
  • Data items to be coded – Including coding dictionaries and versions will be used, specifying if the same dictionary version will be maintained during the course of the study.
  • Country/Site – Specifying the number of countries and sites taking part in the study, including any special considerations/process put in place for a country/site, e.g. initials not to be collected as per a country’s data protection laws.
  • Output documents expected – To include under each section as relevant the document to be created for each task, including a template with the DMP appendix.
  • Responsibilities –  A table of responsibilities including function/role vendor should be included, ideally with start and end dates.

In summary, once a study has completed the set-up phase and all specific study documents [Data Transfer Specification, Data entry/EDC completion guideline, etc] are finalised, the DMP should be a complete reference document for the DM. The time should be spent to maintain the document, updating key information during the course of the study. A key requirement of Good Clinical Practice (GCP) is the documentation of what has happened during the course of a study, the DMP can be the main player in providing this together with the Data Management Report (DMR) to provide evidence of what was planned at the beginning of the study and what was achieved at study completion.