Data sharing – why, when and how?

Aim and intended outcome 

The purpose is to develop an understanding of the issues involved, to share experiences and some practical step that can be taken to facilitate data sharing .       

Introduction  Many research studies in primary care are funded using public money and funders now expect that data collected should be fully exploited, not only by the original researchers but by the wider research community. Data may be requested for the following reasons  Pressure from journals to release individual level data in order to reproduce study results for verification.   To explore new hypotheses, or to assist planning new studies.   To combine the data with other studies in an Individual Patient Data meta analysis   Freedom of Information Requests.   Such requests may be outside the original consent conditions and many researchers are unsure about how to respond to these and what data can be shared while still complying with the Data Protection Act. Researchers may also want to retain sole access to data that they have not had time to analyse.    The workshop will focus on data collected directly by researchers and rather than access to data held by NHS Digital, although some principles may be common to both types of data.    Sally Kerry is chair of the Data Sharing Committee at the PCTU and a member of the Participant Data Sharing Group of the UKCRC Registered Clinical Trials Units. Chris Newby is a member of the Asthma UK Centre for Applied Research, and has experience of statistical methods for data sharing where anonymisation is not possible.   

Format 

After brief introduction a mixture of small group discussions and feedback   

Content 

Introduction – formal talk (10 mins)  Principles of data sharing as a balance of risks between the risk of a participant being identified and the ethical obligation to demonstrate the reproducibility of a study’s findings or to make the best use of data that has been collected.  

Participants introduce themselves and their own learning objectives (10 mins)    Break into groups to discuss different aspects of data sharing with groups formed according to interest and experience

1. Individual patients data meta analyses

2. Consent issues and anonymisation

3. Requests from journals for data sharing

4. What can be done to make data sharing easier (20 mins)   

Feedback from groups (20 mins)    What is anonymisation and consideration of a data sharing request? In small groups participants will look at a data sharing request and discuss whether or not the data could be considered anonymised (20 mins).   

Summing up (10 mins)      

Intended audience 

The workshop should appeal to a wide range of researchers , from novice researchers to Principle Investigators of large studies.