15. When we consider a complaint, we look at whether there are signs the organisation has got something wrong. We do this by comparing what should have happened with what did happen. We have done this and have not found any indications that something has gone wrong.
16. We have carefully considered Mrs C’s concerns and are grateful for the time and work she has put into providing us with a clear and detailed complaint. We thank her for this.
17. Mrs C has asked us to consider whether the MHRA followed its own procedures and acted in line with relevant policies and standards when it made changes to the requirements for sodium valproate.
18. Regulation 26 of the Human Medicines Regulations says the licensing authority (in this case, the MHRA) may at any time suspend, revoke, or vary a manufacturer’s licence, or a wholesale dealer’s licence if it appears to the authority to be necessary to do so for the purpose of safeguarding public health. This is the legal power which allows MHRA to impose safety measures regarding use of medicines when new risks are identified with them.
19. When the MHRA assesses a safety concern, it must follow a structured process as outlined in the PACA guidance note. Up to January 2024, it had to follow the process as outlined in the Assessment Guidance. We have outlined these steps below.
20. The PACA note explains that since January 2025 the process starts by identifying a potential problem with a certain medicine, this could come from patient reports, published data, clinical studies or updates from other regulators. Once the concern is noted, the MHRA must check whether it is valid and serious enough to warrant further assessment. We will refer to this a stage one for the purposes of our decision.
21. If it confirms the concern is valid, MHRA moves to a second stage and takes a close look at the evidence. It must critically assess the information it has received. For example, it will check how well the clinical studies were designed, whether there is a strong link between the medicine and the issue and if the concern make sense. It also compares findings across different sources and considers how relevant the issue is to real-world clinical practice. The PACA note emphasises the importance of using established evaluation tools and frameworks to ensure the robustness of the assessment.
22. Next, at the third stage, MHRA reviews whether the safety concern changes how the benefits of the medicine compare to its risks. It looks at how serious and how common the adverse effect is, whether there are other treatments that could be used instead, and how the issue affects different groups of patients. This helps MHRA decide if the medicine’s guidelines require further action.
23. After reviewing the benefits and risks, MHRA progresses to the fourth stage where it must determine the appropriate regulatory action. The guidance outlines possible actions for MHRA to take which include:
• changing the terms of the medicine’s licence • putting safety measures in to reduce risks • suspending or cancelling the medicine’s licence altogether • sharing important safety updates with healthcare professionals and patients.
24. The PACA note says MHRA must also make sure that stakeholder engagement is part of the process. This means talking with the company that holds the licence for the medicine, as well as medical experts and patient representatives where appropriate. This is the fifth stage of its process.
25. Lastly, for the sixth and final stage of its process the MHRA must establish a plan for follow-up and monitoring. This includes checking how well its actions are working, looking at any new data that comes in, and updating the risk assessment if needed.
26. Having reviewed the Assessment Guidance, we can see during the events of this complaint the MHRA was obliged to take steps two, three and four of the process set out above. The Assessment Guidance did not set out the MHRA needed to follow the other steps that are now included in the PACA note.
27. We have reviewed the evidence we have received to see if MHRA correctly followed the steps outlined above, with reference to the Assessment Guidance in place at the time.
28. Upon review of the evidence, we can see MHRA followed the Assessment Guidance when reviewing the safety of sodium valproate. To provide further reassurance to Mrs C, we can explain we have also seen it acted in line with the additional steps set out in the PACA note, even though they had not yet been formally incorporated into the MHRA policies at that time.
29. We can see the concern about the medicine was raised through multiple sources. L’Agence nationale de sécurité du medicament et des produits de santé (ANSM), the regulator for ensuring safety and quality of medicines in France, started a formal review of sodium valproate in Europe in 2017. It was concerned about the use of the medicine in women who were pregnant or could become pregnant. It said it posed a high risk of birth defects and developmental problems in babies.
30. As the issue affected multiple countries, ANSM requested a formal European-wide safety review, (which is also known as an Article 31 referral) through the European Medicines Agency (EMA). EMA’s safety committee, the Pharmacovigilance Risk Assessment Committee (PRAC), investigated the evidence, held a public hearing, and recommended stricter measures. These included banning valproate in pregnancy (except for rare epilepsy cases), launching the PPP, and updating warnings and packaging. The MHRA has clearly explained how it considered these developments in reaching its own decisions about use of valproate. Therefore, although this was not a step indicated in the Assessment Guidance, we can see MHRA identified an issue to explore. This would be in line with the first stage of the PACA guidance.
31. Following this, MHRA took further action to check the concern was valid and to critically appraise the evidence it had received. We can see from the MHRA public assessment report it gathered further evidence from prescribing data and medical records from women who had been pregnant and exposed to sodium valproate. Its report found that women of childbearing age were still being prescribed the medicine and becoming pregnant. It obtained the data from the UK Medicines and Pregnancy Registry and the Clinical Practice Research Datalink (CPRD). This data independently confirmed the concern about sodium valproate was valid and serious enough to warrant further assessment. MHRA also received feedback from patient groups and healthcare professionals who raised concerns about poor compliance with the PPP.
32. Once it obtained evidence to show the concern about sodium valproate was supported, MHRA carried out a detailed review of the evidence. We can see in its public assessment report this included clinical studies and research which included looking at the known risks, but also newer concerns such as the effects on male fertility and possible long-term genetic changes. MHRA used data from CPRD to understand how the medicine was being prescribed and whether patients were being switched to safer alternatives. MHRA also looked at data from France to compare patterns.
33. Alongside the clinical evidence, we can see in its public assessment report MHRA also gathered feedback from patients, clinicians and expert groups. To make sure its review was thorough, MHRA applied a structured approach and used tools to assess how reliable the studies were. This included looking at whether the studies included enough people to produce reliable results, if the researchers used the right tools to analyse the data and if the studies were well designed to support evidence-based decisions which would be applied in everyday clinical care. It also looked to see how strong the link was between valproate and the risks, and how relevant the findings were to everyday clinical practice. We consider this action was in line with the obligations set out in the Assessment Guidance and the second stage of the PACA note.
34. We can see MHRA then took steps to weigh the benefits of valproate against its risks. The public assessment report shows it looked at how serious the side effects of the medicine were, how often they happened, whether safer alternatives were available for use and how different patients may be affected.
35. MHRA made a clear distinction between epilepsy and bipolar disorder. For bipolar disorder, it said valproate should not be used during pregnancy. For epilepsy, MHRA allowed it only if no other treatment worked or was tolerated. This decision is clearly explained in the public assessment report.
36. Within the MHRA public assessment report we can also see it considered what it may mean to switch a patient’s medication from sodium valproate, especially the risk of losing seizure control or causing other harm. MHRA emphasised that any changes should be led by specialists who understand the patient’s needs. From the evidence we have seen, we consider MHRA acted in line with the Assessment Guidance and the third stage of its PACA note.
37. After reviewing the benefits and risks of valproate medicines, the MHRA took regulatory action as outlined in the Assessment Guidance and the fourth stage of its PACA note. As we have mentioned, in 2018 EMA had already introduced stricter safety measures, including the PPP. At that time, the UK was still under European law, so MHRA was responsible for putting those changes into effect.
38. By 2022 and 2023, MHRA had gathered new evidence as we have seen in its public assessment report. The report found evidence existing measures were not working well enough.
39. In response, in January 2024 MHRA introduced further restrictions on valproate medicines. These included a requirement that no one under 55 should start valproate unless two specialists agree there is no safer alternative. Patients already taking valproate must also be reviewed by two specialists to confirm that continued use is appropriate. The MHRA also updated the medicine’s license and included its restrictions on the patient guide for valproate.
40. Within its public assessment report, we can also see the MHRA refreshed educational materials on sodium valproate and rolled out a phased implementation plan to avoid disrupting patient care. We consider these actions are in line with the Assessment Guidance and PACA note, which include changing the licence, adding safety measures, restricting use and sharing safety updates.
41. Although not a requirement set out in the Assessment Guidance, we can see the MHRA appropriately engaged with stakeholders as outlined in stage five of its PACA note. This is because we can see evidence in the EMA Valproate referral that the EMA held a public hearing in 2017 and invited written input from patients, doctors, and regulators. Following this, we can see evidence in its public assessment report that MHRA also consulted with patients, charities and clinical experts and considered evidence from the Valproate Implementation Expert Working Group, which included NHS leaders and specialist societies. This group helped shape the rollout and educational materials, showing that MHRA involved the right people throughout the process.
42. Lastly, in the MHRA public assessment report we can also see evidence it acted in line with the final stage of its PACA note as it committed to follow up and monitoring of the changes it has made to sodium valproate medicines. It provided evidence that it reviewed how well the PPP was working and found it was not effective or consistent. The MHRA used data to track prescribing and reviewed French data to compare trends. The MHRA has also committed to reviewing new data, including possible risks from patient exposure. This shows the MHRA is checking how well its actions are working and continues to update its risk assessment, as the PACA process from January 2024 requires.
43. With this in mind, we have not seen anything to indicate that the MHRA failed to act in line with applicable guidelines and standards in the way it reached its decision. We explored why Mrs C feels so strongly that there must be maladministration in the MHRA’s process.
44. Mrs C has told us that because MHRA has changed the requirements for sodium valproate she no longer has access to the medicine she needs. We understand this must be extremely upsetting and worrying for her. After carefully reviewing the evidence, we have not seen that the MHRA’s regulatory decision means patients who have a clinical need for valproate are unable to obtain it. We recognise MHRA has made it more restrictive to access sodium valproate medicines, but it also made it clear in the updated changes that it can be prescribed in certain circumstances, as we have described in paragraph 36. If Mrs C has concerns that she is not able to access medications for which she has a clinical need, it is open to her to discuss those concerns with her GP or other appropriate clinicians to agree an way forward, and we hope any such discussions are helpful to her.
45. In conclusion, we are of the view MHRA carried out changes to sodium valproate medication in line with its role and obligations and have therefore not seen indications anything went wrong.
46. As we are satisfied MHRA acted in line with the guidance we have mentioned, we will take no further action. We know these matters are important to Mrs C, so we hope that we have clearly explained above how we reached our decision in her case.