Dr. Almath Spooner:

The two main questions addressed to us were from Deputy O'Connell. My understanding is that there were two main pillars to those. One was on the risk minimisation. Communication does not seem to have been optimally delivered and the Deputy asked what we are doing differently now. We are making it a requirement of the licence that these things are done. I refer to any failure to provide patient cards or failure to provide precise information on the magnitude of the risk. Use of the medicine in a woman of childbearing potential who is not on effective contraception will be off licence now.

That is a change.

In terms of the enablers, we are putting a formal pregnancy prevention programme in place. That is recognising that there have been deficiencies associated with a lack of clarity on different roles and responsibilities around the patient journey. What is the role of the regulator, the specialist, the general practitioner and the community pharmacist? What are the clinical enablers that allow steps to be taken at the appropriate time in terms of decisions on therapeutic options but also information provision? The new recommendations and the new clarity on the licence, supported by what will be a pregnancy prevention programme that will be branded "Prevent" and that will have a suite of measures that have been fully informed by engagement with patients and health professionals, will be a very substantial change and a framework around risk communication as opposed to ad hocbulletins and the like. We have to recognise that we are doing this for this medicine because of the magnitude of the risk, which is not neutral in terms of burden. This is a medicine a woman will be on for a long period of her life and the risk of pregnancy will vary, so it needs to be adapted to the woman's individual circumstances.

In terms of moving forward, we have a more robust approach now to communication, at least from the perspective of what we are accountable for, which is the regulation of the product.

On the question about the other pillar of risk management, which is around evidence generation and how we research medicines at a population level when we have uncertainties, linked to that is the availability of electronic resources, electronic health records, and having the possibility to research population use of medicines. In terms of much of the pharmaco-epidemiological research, we have had a lot of discussion on the complexity around interpreting individual cases but we have had studies which can get around many of those issues with sophisticated data linkage. That is not always possible for every medicine but what is clear now is that for every medicine we have a risk management plan and at the time of initial authorisation we ask about the uncertainties and we plan proactively to reduce those uncertainties. If sodium valproate was being put on the market today, having seen the non-clinical data and having the concerns about physical defects, we would be requiring the company to do pharmaco-epidemiological studies.

I hope I can provide some reassurance that, going forward, lessons have been learned. We have a much more robust approach to the vigilance of medicines in pregnancy. Perhaps the thinking in the past was that we cannot research medicines that are used by women who are likely to become pregnant. There was a reluctance to research. That has evolved. We understand that women will be treated for chronic conditions. They will want to plan pregnancies and they need reliable and complete information on the relative harms for different medicines in pregnancy. As regulators, we are accelerating efforts, a term used by Deputy O'Connell, and that is a fair reflection. As part of that, and I do not want to get into regulatory speak, in the next year at the Pharmacovigilence Risk Assessment Committee, PRAC, as well as having the research we are doing at national level, we will be developing a good vigilance practice guideline that will be binding on the marketing authorisation holders. That will be taken into account in all of the risk management plans and the planning of studies to reduce uncertainty so that we do not see these kinds of issues arise in the future with such a time lapse before we get complete and accurate information on the magnitude of the risk. I hope that answers the question but if I have overlooked anything, please remind me.