Intended for healthcare professionals

Rapid response to:


Pandemrix vaccine: why was the public not told of early warning signs?

BMJ 2018; 362 doi: (Published 20 September 2018) Cite this as: BMJ 2018;362:k3948
cropped thumbnail of infographic

Infographic available

Adverse events: GSK pandemic influenza vaccines

Read all The BMJ's investigations

Rapid Response:

Re: Pandemrix vaccine: why was the public not told of early warning signs?

Dear Sir,

We read your article which states that H1N1 pandemic vaccine safety monitoring was not transparent. Whereas the motivation for the paper is clearly narcolepsy litigation, you seem to cast doubt on the entire process of vaccine pharmacovigilance, which in our opinion is unbalanced and may cause needless public anxiety.

Unfortunately, all medicines have undesirable effects, and while vaccines are amongst the safest, adverse events do occur. Not all adverse reactions can be predicted, and narcolepsy (a very rare disease) had never been associated with vaccines before: the signal came as a complete surprise in August 2010.

Narcolepsy case reports were only reported to the health authorities from the summer of 2010 onwards [1]. Narcolepsy experts in Finland did not immediately report them to the authorities [2]. It is important for the public that this information was not withheld by EMA, nor GSK. In fact when the concern was raised in August 2010, immediate action was taken by EMA and each action was reported publicly (see EMA website). Much effort has been made to investigate and understand the unexpected increase of narcolepsy cases following Pandemrix vaccination. As of yet it is unproven what the mechanism was, it seems that H1N1 virus itself may increase narcolepsy, based on data from China and the SOMNIA study in Taiwan [3]. After searching for a mechanism for 8 years, the only plausible hypothesis seems an interaction between H1N1 wildtype virus infection and the administration of an adjuvanted vaccine. Although it intended to be investigated in the recently published SOMNIA study, there was not have enough power to test it [3,4,]. In a workshop organized by International Alliance for Biological Standardization in Brussels March 2018, on the narcolepsy – pandemic vaccine issue, Prof. PH Lambert described the biological basis for this interaction which he called the double hit hypothesis [Lambert PH, personal communication]. Understanding this potential interaction is important for future pandemic preparedness, where adjuvants may need to be used, to get a proper response in vulnerable patients that do not have a proper immunological response yet/anymore, or to spare antigens.

Our main concern is that the paper plays on emotions of fear and anger of the public, fostering mistrust and suspicion, while being incomplete in presenting the full evidence and facts that are publicly available. Spontaneous reports as the authors base their conclusions on, serve to signal, but have many well-recognized limitations, and cannot be just compared between vaccines, which have different roll out periods and target populations. For proper product comparison, epidemiological studies are needed, which were conducted, but not referenced. We live in a period of heightened vaccine hesitancy especially in Europe, in which many parents have a lack of trust. We may have forgotten about the positive impact vaccines have made in our world but we are facing measles outbreaks and deaths again in Europe, which could have been avoided by vaccination. We need to step up to face vaccine hesitancy [5]. As part of the stepping up we would expect BMJ to give a scientifically and evidence based opinion as it impacts on many healthcare professionals who need to talk to vaccine hesitant parents.

The public and healthcare professionals deserve to know that the time of regulatory approval of the pandemic influenza vaccine was shorter, not because of less rigor but because regulators were prioritizing this work in line with the pandemic crisis guidance (publicly available) with larger teams and rolling reviews of available documents to allow for timely licensure, and protect public health. Pandemrix, Focetria and Celvapan were licensed faster because of rigorous mock-up data, where the vaccine construct (including the adjuvant) is already tested, but only the strain is replaced. EMA has been very open about the process and steps taken. Pre-licensure safety information is always limited and will never pick up serious rare events, that is why we have a post-marketing system. Also, mock-up vaccines are not studied in vulnerable groups, such as children and pregnant women, simply because the risks outweigh the benefits for experimental studies and no infectious threat, which is different from when there is a real public health hazard.

We all fully agree that valid and timely information needs to be supplied to the public. In fact, EMA did this throughout the pandemic on a weekly basis until April 2010, in an unprecedented transparent effort and had specific pharmacovigilance teams looking at the reported cases.

As part of the lessons learned and the experience in the ECDC funded collaborative multisite epidemiological studies on Guillain Barre Syndrome and narcolepsy to quantify associations, it was established that Europe should have a sustainable platform for multinational epidemiological research which would allow for rapid quantification of safety signals. The Innovative Medicines Initiative funded a public private consortium in 2013 that engaged 47 partners throughout 19 countries in the EU, with the vision of transparently and collaboratively monitoring the benefits and risks of vaccine using available health data. Such a system has been built and tested, and its deliverables are freely available ( It will continue in 2019 as an not-for profit association: VAC4EU aiming to have a lasting European system for actively monitoring vaccine coverage, benefits and risks which could help to support decision making by providing the evidence.

We feel that it would serve public interest to get a proper view on the level of transparency of pharmacovigilance in the EU. While seemingly partial presentation of facts may be accepted for sensationalist newspapers, it should not be the standard for scientific journals. As society and scientists, we should use our energy to further improve big health data collaborations that have demonstrated the ability to provide post-licensure safety data in a timely and transparent manner in order for regulators and healthcare professionals to have proper evidence when they speak with parents.

1. Gadroen K, Straus SMJM, Pacurariu A, Weibel D, Kurz X, Sturkenboom MCJM. Patterns of spontaneous reports on narcolepsy following administration of pandemic influenza vaccine; a case series of individual case safety reports in Eudravigilance. Vaccine. 2016 Sep 22;34(41):4892-4897.
2. Accessed 25 September 2018
3. Weibel D, Sturkenboom M, Black S, de Ridder M, Dodd C et al. Narcolepsy and adjuvanted pandemic influenza A (H1N1) 2009 vaccines – Multi-country assessment, Vaccine 2018; 36: 6202-6211
4. Sturkenboom MC. The narcolepsy-pandemic influenza story: can the truth ever be unraveled? Vaccine. 2015 Jun 8;33 Suppl 2:B6-B13.
5. Coombes R. Europe steps up action against vaccine hesitancy as measles outbreaks continue BMJ 2017;359:j4803

Competing interests: MS has conducted several safety assessments of the pandemic vaccines funded by CDC and ECDC. She is coordinating the ADVANCE consortium. SB has a contract with Novartis on a topic unrelated to this paper and was principal investigator of the SOMNIA study

26 September 2018
Miriam C Sturkenboom
Prof. Steve Black (University of Cincinnati, USA)
Julius Global Health, University Medical Center Utrecht
Heidelberglaan 100, Utrecht, The Netherlands