Swine flu jab narcolepsy risk is very small

There have been reports in the media today that the Pandemrix vaccine, used during the swine flu epidemic of 2009-10, increases children’s risk of developing narcolepsy.

Underpinning the news is a well-conducted, government-funded study that confirms findings of previous research from Finland.

However, it is important to point out that the risk of narcolepsy – where a person suddenly falls asleep at inappropriate times – is thought to be extremely low. The researchers estimate that the chance of developing narcolepsy after receiving a dose of the vaccine is somewhere between one in 52,000 and one in 52,750.

Due to the results of previous studies, Pandemrix is no longer given to those under 20. There is no evidence that other types of vaccine carry a risk of developing narcolepsy.

Children are not routinely vaccinated against flu, although vaccination was extended to children during the swine flu pandemic. Being vaccinated against flu is very important for over-65s and others at high risk of developing complications from the condition.

Where did the story come from?

The study was carried out by researchers from the UK Health Protection Agency, London, Addenbrooke’s Hospital Cambridge, University College London and Papworth Hospital, Cambridge. It was funded by the Department of Health and the Health Protection Agency.

The study was published in the peer-reviewed British Medical Journal and has been made available on an open-access basis so it is available for everyone to read.

The coverage in the media was mostly accurate and took a responsible tone by explaining that the risk is very small. The lessons of the MMR scare (which turned out to be based on discredited evidence) appear to have been learnt by the media, researchers and medical journals alike.

However, the Daily Mail’s headline statement that there are “fears one million children received jab” is perhaps poorly worded as it exaggerates the scale of the potential risk. If one million children did receive the Pandemrix vaccine (which is a matter of debate), even at the highest risk estimation (one in 52,000), just 19 children would be expected to develop narcolepsy.

What kind of research was this?

This was an analysis of the cases of children and young people in England who developed narcolepsy from January 2008. The aim was to evaluate the risk of narcolepsy associated with the Pandemrix vaccine.

The authors point out that narcolepsy is a chronic sleep disorder thought to affect between 25 and 50 people in 100,000, with onset most common in those aged 10-19. It is characterised by excessive daytime sleepiness, often accompanied by temporary loss of muscle control triggered by strong emotion (cataplexy).

The Pandemrix vaccine against swine flu (also referred to as H1N1 influenza because it is the H1N1 strain of the virus) was introduced in England in October 2009, during the global H1N1 pandemic of 2009-2010.

It was given to children under five from December 2009 and by March 2010 nearly a quarter of healthy children under five and 37% of those aged 2-15 with conditions that made flu a high risk, had been vaccinated. A second vaccine called Celvapan was also used but accounted for less than 1% of the total.

In August 2010, concerns were raised in Finland and Sweden about a possible association between narcolepsy and Pandemrix, and a study in Finland found a 13-fold increased risk of narcolepsy after vaccination in those aged 4-19.

The current study was set up to evaluate the risk of narcolepsy in England after vaccination.

What did the research involve?

Between 2011 and 2012, the researchers collected data from 23 sleep and paediatric neurology centres in England, on children and young people who had developed narcolepsy between the ages of four and eight, from January 2008. 

As an alternative method of ascertaining cases of the disorder, they also identified all cases recorded in a hospital statistics database.

The patients’ GPs were contacted to find out:

  • whether they had been vaccinated against seasonal flu and swine flu
  • the date of onset of narcolepsy symptoms
  • date of first consultation
  • date of referral for sleep problems
  • information on any infections before narcolepsy onset

The cases were all reviewed by an expert panel to confirm the diagnosis, according to international classification of sleep disorders criteria. Cases that did not meet the criteria were classified as ‘probable narcolepsy’. The panel were ‘blinded’ to the children’s vaccination status (they did not know which children had been vaccinated).

In their statistical analysis, the researchers compared the chance that vaccinated children were seen for narcolepsy with the chance that a control group of unvaccinated children of the same age and with the same risk group status were referred for the condition.

The data for the control group came from a case-coverage analysis. This is a technique used for analysing rare adverse events after vaccinations. As it requires only a sample of the cases, it avoids having to follow large population cohorts or selecting controls.

For estimating the overall number of people eligible for vaccination, the researchers relied on GP register data.

What were the basic results?

After reviewing case notes for 245 children and young people with narcolepsy, the researchers found that 75 had developed the condition after January 2008. Eleven children and young people had been vaccinated before the onset of symptoms. Of these, seven had been vaccinated within the six months before onset.

For the control group population data, they extracted information on 160,400 individuals aged between two and 18 from GP records.

Their analysis found that vaccination at any time was associated with a 14–fold increased risk of narcolepsy (odds ratio (OR) 14.4, 95% confidence interval (CI) 4.3-48.5), and vaccination within six months of the onset of symptoms was associated with a 16-fold increased risk (OR 16.2, CI 3.1-84.5).

They calculated the vaccine-attributable risk of narcolepsy to be between one in 57,500 and one in 52,000 doses. This means that about two children per 100,000 could have been spared narcolepsy if they had not been vaccinated.

How did the researchers interpret the results?

The researchers say they found evidence of an increased risk of narcolepsy in children who received the Pandemrix vaccine in England, similar to that found in the previous study in Finland. They say their research indicates that the vaccine can cause narcolepsy.

However, the authors rightly say it is possible the risk has been overestimated. Children with narcolepsy can go undiagnosed for many years as there are many possible causes of their symptoms. It may be possible that children who would have developed narcolepsy – whether they were vaccinated or not – were referred and diagnosed sooner because of increased public awareness of a possible link. This may have slightly skewed the results of the study.

The researchers say the findings have implications for the future licensing of similar pandemic vaccines. Further studies are needed to assess any risk associated with other vaccines developed against swine flu.


This study confirms that the Pandemrix vaccine against swine flu is associated with a very small risk of narcolepsy in children and young people.

As the authors point out, this risk may have been overestimated if children with narcolepsy who had been vaccinated were more rapidly referred than others because of increased awareness of the link.

The methods used were practical for a rapid assessment of risk, but as this was essentially a case series analysis they are limited by a number of factors:

  • The rates calculated are dependent on the accurate diagnosis and identification of the cases of narcolepsy. By using a particular diagnosis code (for narcolepsy and cataplexy) to identify hospital admission alongside the case reports it is possible some eligible cases were missed from the analysis.
  • Out of the 23 centres asked only 16 replied that they had seen affected children in the relevant period and provided data.
  • Baseline incidence was hard to estimate and capture, it varied widely between countries. There were also apparently significant increases and decreases in incidence in individual countries unrelated to vaccine use.

There will inevitably be concerns that the Pandemrix vaccine was distributed too quickly and further testing should have been carried out. This ignores the context of the threat posed by swine flu at the time. At the height of the swine flu pandemic, millions of cases were occurring across the globe and there was real uncertainty about the public health impact the epidemic could have.

A decision was taken to speed up use of the vaccine and, as always, this was a judgement that carefully weighed up the risks and the benefits. Most experts would agree that the benefits of vaccinating children against H1N1 swine flu outweighed the very small risk of narcolepsy.

NHS Attribution