= SCAD and CAE

Extension of the [[SCAD Study]] to include patients with '''Coronary Artery Aneurysms / Ectasia''' funded by [[http://www.nihr.ac.uk/about/rare-diseases-translational-research-collaboration.htm | Rare Diseases Translational Research Collaboration]] ([[http://rd.trc.nihr.ac.uk/|Website]]).

- [[attachment:Study guidance.doc|RD-TRC Study Guidance]]
- [[attachment:NIHR RD-TRC Data Day Agenda (pm) 28th Jan 2015.docx|RD-TRC Data Day Agenda]]

== Data Meeting

Roger James of the [[http://diligenti.net/rdcit.org/|Rare Diseases Clinical Infrastructure Team]] (RDCIT) gave an introduction to OpenClinica and data requirements for the RD-TRC.

=== Data Requirements

- There are no specific data requirements for the RD-TRC, except that patient identifiable study data is stored in the RDCIT instance of OpenClinica by the end of the funding period (approx September 2016).
- There is a [[attachment:GEL minimum data spec.xlsx|proposed data structure]] that is based on the data requirements of GenomicsEngland (GEL).  This includes patient identiable details, consent, medical / family history, genetic / clinical tests, medication and more.  There is, however, no formal link between RD-TRC and GEL and all fields that are mandatory in GEL are not mandatory for RD-TRC.
- To this we can create a data definition (CRF) to facilitate our study data.
- These data definitions (CRF) are published on the RDCIT OpenClinica site and reuse and collaboration are encouraged.

=== OpenClinica useage

- OpenClinica allows you to define sets of study data to be entered by clinicians.
- The definitions are recorded in Excel spreadsheets and are version controlled.  So that when data is extracted you can tell the particular version of a question the clinician was asked and what options they were give.
- Data can be entered manually into OpenClinica or it can be imported from a spreadsheet.
- The import process requires you to map the column in the spreadsheet to a particular version of a question in a CRF.  (In the demonstration the column in the spreadsheet was renamed to be specific question ID from the CRF XML definition.  I don't know if this is necessary or helps with the mapping.)  The mapping can then be saved and reused with many spreadsheets.
- Importing seems to be done in 2 stages.  The first stage takes the input file and converts it to several mapped files. (So as not to be too large?)  The mapped files are then imported.

=== Infrastructure

- The RDCIT instance of OpenClinica currently sits on a server in Cambridge this instance is '''not secure for storing Patient Identifiable Data''' and so is called OC Anonymous.
- RDCIT have commissioned a server in a data centre in Liverpool that will comply with various security requirements and so will be suitable for patient identifiable data.  This will be called OC Secure and should be ready in June.
- Both servers are visible on N3 and the equivalent university network.

=== Remaining Issues

After the presentation several questions and issues were raised, some of which require further clarification:

- Ethics and Consent: It may be necessary to re-consent patients to put their data on the RDCIT OpenClinica instance.  GEL has ethics policy that should be reusable for this project.
- Data Sharing: There seemed to be no consensus on whether Data Sharing Agreements would have to be created for all involved organisations.  It was put forward that as long as a responsible clinician entered or approved each individual data entry, it should be OK.  Whether the Trsusts would be OK with this is another matter.  RDCIT was going to seek clarification from the Department of Health.
- Responsibility for the coding and mapping of the data lies with the study team, although technical assistance may be available from RDCIT through the Theme coordinator.
- Funding for IT should be covered by the BRU budget as this is an amendment to the BRU contract.
- Requests for new users, groups and permissions should be made to RDCIT.  There is no devolved user / permissions admin.