Acceptability of linking routinewly collected NHS data sets to create a paediatric pharmacovigilance database

Introduction
There is widespread recognition of the vulnerability of children to adverse drug reactions (ADRs) 1,2. Paediatric pharmacovigilance is a recognised priority. The UK Yellow Card Scheme (YCS) is central to medicine surveillance, but other methods have been suggested as useful adjuncts3. The inclusion of the community health index (CHI) in all NHS contacts in Scotland will facilitate linkage of multiple datasets on a routine basis to create a database which could be used for, for example, pharmacovigilance. However there are questions about the acceptability, and legal and ethical implications of this. We are engaged in a CSO funded programme of work to address these issues and develop novel methods of early ADR detection in children. The aim of the work reported here is to assess health care professionals’ opinions of, and attitudes to, current systems of pharmacovigilance, advantages of, and barriers to, the national linking of data, and to propose solutions.
Methods
Semi-structured interviews are exploring views on pharmacovigilance, confidentiality/patient privacy, data protection, acceptable and non- acceptable usage of data, and dissemination of findings. A purposive sample of national experts, REC chairs and Caldicott guardians from all Scottish Health Boards are participating. Ethical approval was granted (NoSREC).
Results
Analysis of the early interviews (n=15) demonstrates awareness of the YCS but identifies limitations: “it [spontaneous reporting] doesn’t produce data, it produces anecdotes” (A04). Practical, ethical and legal concerns are raised. Practical issues include physical security of data and anonymisation: “anonymisation [...] is not a science, it’s a craft” (A11). The need for consent is discussed widely, a blanket approach seen as “not feasible” (A) but public awareness perceived as crucial: “[the] patients need to be aware of what is proposed....... [the] use the data would be put to” (A05). Legal issues are related to Acts, professional standards and codes of conduct.
Conclusion
Data linkage is seen as beneficial but concerns are raised. These results cannot be quantified or generalised. They will inform a Delphi survey to be distributed to a national sample of professionals. The outcome will be agreed solutions to inform the next stage when datasets will be linked.

References
1) Ekins-Daukes S, Helms PJ, Simpson CR et al (2004). Off-label prescribing to children in primary care: retrospective observational study. Eur J Clin Pharmacol 2004; 60: 349-353
2) Choonara I, Conroy S (2002). Unlicensed and off-label drug use in children: implications for drug safety. Drug Safety 2002; 25: 1-5
3) UKCRC, 2007. UKCRC R&D Advisory Group to Connecting for Health: The Report of the Research Simulations. Available at: www.ukcrc.org/publications/report [accessed 04/03/2010]