The 2009 H1N1 swine flu pandemic: don’t panic but you are all going to die
This essay was written by Peter Coombs and was first published in the 2010 Mill Hill Essays. An updated version was published in the Mill Hill Essays anthology.
‘Don’t Panic But You Are All Going To Die’ is a headline from the satirical news website The Daily Mash at the time of the swine flu outbreak, published 27th April 2009.
In August 2010 the World Health Organisation (WHO) declared that the ‘swine flu’ pandemic was over. In the preceding 17 months, the pandemic had spread across the world infecting millions of people. Since the announcement of the end of the pandemic the WHO, national health organisations and the media around the world have been questioning whether the response to it, and the associated cost, was warranted.
The first confirmed human cases of the new swine flu were in Mexico in March 2009, where a number of serious cases drew attention to the outbreak. Mexico informed the WHO and the United States, and attempts were made to contain the spread. Later analysis has revealed that there were probably hundreds of non-lethal cases before March; the virus appears to have first passed to humans at the end of 2008. In mid- April 2009 the first cases in the US were confirmed, and by the end of April the first cases in Europe were being reported. On June 11th 2009, the WHO raised its influenza pandemic alert to phase 6 and declared that the virus was causing a global pandemic.
The initial outbreak gave rise to a massive amount of media attention. The tabloid press in the UK did not hold back; ‘Third of world could catch swine flu’ said a Daily Mail headline. Despite the sensationalism of the tabloids, the new outbreak did represent a very real risk. It was a novel strain, so there would be no, or only limited, immunity in the population; the severity of infection and ease of spread of the virus were unknown. Estimates of the numbers who would be infected worldwide and how many would die varied massively, and the mass media tended to focus on the biggest ‘worst case scenario’ numbers. Ben Goldacre in his excellent Bad Science column commented on the difficulty of predicting what would happen with the swine flu outbreak, ‘infectious disease epidemiology is a tricky business: the error margins on the models are wide, and it’s extremely hard to make clear predictions’. The reality was that early on no-one knew what would happen – it could be bad, or it might not be.
Early on in the swine flu outbreak some indicators weren’t good. The virus emerged quite quickly and spread rapidly. It appeared to be infecting significant numbers of young people, which was also a noted characteristic in reports of the 1918 pandemic, which killed tens of millions worldwide. It was also an H1N1 virus strain, the same as in 1918. Initial cases in the United States and Europe were mostly mild but even with a low mortality rate, if enough people were infected, the numbers of deaths could be large and the strain on health services significant. A critical concern was that if the new swine flu strain was to mutate or combine with another influenza strain, then a more virulent and dangerous strain could emerge.
The sensationalism of the tabloid newspapers was only surpassed by the ill-conceived conspiracy theories on the internet. Fuelled perhaps by fear of the unknown, combined with a large dose of imagination, some of the more outlandish rants suggested that the virus had been created as a government ploy to reduce the world’s population, or that pharmaceutical companies had produced and released the virus to make billions of dollars in vaccine and drug sales.
As part of the media hype, particularly a few weeks into the outbreak, another tack was seen with a selection of ‘what is all the fuss about’ stories. One of the most outspoken people was Simon Jenkins writing in the Guardian, who wrote a couple of articles early on in the pandemic saying that the ‘risk to Britons’ health is tiny’. He suggested that the scaremongering coverage was about selling drugs and justifying budgets and was a waste of resources. Several rebuttals from people both within and outside the scientific community commented on the reality of the risk and the need to act rationally in the face of media hype and the necessity of being prepared.
A range of scientists and health advisors gave multiple interviews in the media. The health secretary and the government chief medical officer were among those constantly in the news media explaining what was happening, what the plans were and encouraging people to be aware of the danger but not to panic. Virology experts, including NIMR’s Alan Hay and John McCauley (the past and current Directors of the WHO Collaborating Centre for Reference and Research on Influenza at Mill Hill) gave interviews with daily newspapers, BBC radio and TV programmes, explaining the unusual genetic make-up of the virus, progress in vaccine development and how severe the spread might be. Estimates of the spread and severity of the pandemic were always qualified with the caveat that these were estimates based on available evidence, but that it was unknown how it would progress and we should be prepared.
Public perception of swine flu and the media coverage is likely to have been affected by experience with other recent potential pandemics, particularly the SARS outbreak and the threat of H5N1 bird flu in the early 2000s. That these viruses did not result in worldwide devastation after the surrounding tabloid hysteria has made people more sceptical. Does that mean that the swine flu outbreak was overhyped? Not necessarily. It is sometimes difficult to accept, certainly in the cut-and-dried world of the mass media, that things are unknown. Risks exist everywhere, though particularly for things like influenza pandemics, the severity and impact of a new outbreak are very difficult to estimate early on, and things can change as the pandemic spreads.
It is very difficult to predict the spread and severity of a novel influenza outbreak, or indeed any disease outbreak. Early reports and fatality numbers can be inaccurate and deceptive due to factors such as selection bias (only those with serious problems are observed), media bias (people dying is more likely to be reported than people getting better), and incorrect reporting (for example, the initial estimated fatality rate of the outbreak in Mexico proved to be too high and was later revised downwards). Complications such as delays in reporting and sample testing, non-reported cases, and confusion with other illnesses with flu-like symptoms, further confuse the matter. Yet the scale and type of response needed to combat the flu outbreak would depend on how the virus was spreading and behaving.
Looking at the influenza pandemics from the 20th Century, the mortality rate has varied considerably, ranging from around 1 million deaths in the 1968 Hong Kong flu to an estimated 20-100 million deaths in the 1918 Spanish flu. These figures should be placed in the context that 250,000- 500,000 deaths worldwide are attributed to seasonal flu each year. Influenza is a big worldwide killer, even outside pandemics.
In typical influenza infections it is the very young, the elderly and those with underlying health problems who usually suffer the most. However, with the 2009 outbreak elderly populations only showed low levels of illness. Studies showed that a significant proportion of those in the 60- plus age range had some degree of immunity to the new pandemic strain, either from exposure to circulating strains when they were very young or from earlier vaccinations. This is because the 2009 swine flu virus shares some immune characteristics with the viruses that were around 60-plus years ago.
The part of the population most affected were people less than 30 years old, mostly having mild symptoms. A small proportion of infections during the pandemic resulted in quite severe viral pneumonia; especially among 30-50 year olds. This group of people are not typically a high risk group, though many of those affected had underlying medical conditions.
So what makes the swine flu virus different and where did it come from? The swine flu virus is of the influenza strain H1N1. The H and N refer to the proteins found on the surface of the influenza virus, hemagglutinin and neuraminidase, which are numbered (H1-H16; N1-N9) depending on which subtype they belong to. The hemagglutinin binds to host cells, allowing the virus to enter the cells. The neuraminidase is an enzyme which helps newly produced viruses escape from the host cell. These two proteins are present in large numbers on the surface of the virus, and so are also the parts of the virus that the host immune system will see. When the human body is infected with the virus or vaccinated it will develop antibodies against the virus, and be immune to further infections by the same virus. However, the influenza virus has evolved to regularly mutate these exposed surface proteins, so that the human immune system does not recognise them.
The influenza virus contains 8 gene segments, each of which contains the information necessary to make the proteins that the virus needs to help it infect a cell, to replicate, and for its progeny to escape the host cell and go on to infect other cells. Sequencing of the genes of the 2009 H1N1 swine flu virus led to the finding that the strain is related to viruses recently circulating in pigs in North America and in Europe/ Asia. The strain is made up of a mixture of these genes, for example the hemagglutinin is descended from the North American ‘triple reassortant’ strain and the neuraminidase from recent European/Asian swine viruses. The strain itself had not been detected previously in human or swine populations. The ‘triple reassortant’ viruses from North America have been in circulation in pigs since the late 1990s, and contain genes from older avian, human and swine viruses.
Pigs have been termed a mixing vessel for influenza viruses because they can be infected by both avian and human influenza viruses. If pigs become infected by more than one virus at the same time, the viruses can swap genes producing new variants, potentially producing a strain that is transmissible to and by humans.
Drugs and vaccines
Since the H5N1 bird flu outbreaks in South-East Asia in 2004 onwards, the UK government had been stockpiling the antiviral drugs Tamiflu and, to a lesser extent, Relenza for use in the event of a pandemic. Tamiflu and Relenza, also known as oseltamivir and zanamivir, work by blocking the neuraminidase on the virus surface, so that new viruses that are produced can’t escape from the host cell. The drugs are effective in reducing the length and intensity of flu infections if taken within the first few days of contracting the virus. This stockpile in the UK amounted to 30 million doses, enough for half the population, and more were ordered when the pandemic arrived.
A few years ago, a mutation in neuraminidase was discovered which makes the virus partially resistant to Tamiflu. Work conducted by scientists at NIMR has shown how and why this happens. In the pre-2009 seasonal H1N1 influenza virus, there was a high occurrence of this mutation in some areas of the world. There was concern that this mutation might occur in the swine flu pandemic virus and spread. This led some countries to increase their stocks of Relenza, to which the Tamiflu-resistant strains were still susceptible. Tamiflu is generally preferred to Relenza for treatment because Tamiflu is taken as a tablet, whereas Relenza is taken using an inhaler. Fortunately, only isolated cases of Tamiflu-resistant strains were found during the pandemic.
A national flu service was established in the UK to reduce the strain on hospitals and GPs, and a system for providing antiviral drugs (Tamiflu and Relenza) to people who needed them was set up. The Health Protection Agency (HPA) and the NHS, as well as the WHO, provided guidelines and information on what to do if you were infected and how to minimise the spread of the virus.
Each year the UK, and many other countries, enact a large scale influenza vaccination scheme targeting the currently circulating seasonal influenza strains. The seasonal influenza vaccine was found to give very low protection against the new swine flu strain. The early decision to produce a vaccine for the swine flu for the northern hemisphere winter 2009- 2010 was influenced not just by the unknown severity of the new virus, but given that the virus appeared to be spreading worldwide even if it was mild, the lack of protection from the seasonal vaccine would be like not having a vaccine at all, which would result in many more deaths, particularly in the young, elderly and infirm.
By mid-May 2009 the UK government had agreed with vaccine producers that a separate swine flu vaccine would be produced if the WHO pandemic level was raised to 6, with enough doses for the whole population. A high-yield virus is needed for producing vaccine, where the virus grows in large amounts in the chicken eggs used to produce it. Candidate vaccine strains were developed that grew well, followed by safety testing and scaling up. However, this process takes several months, meaning the first batches of vaccine were available in the UK only by late autumn, separate from the seasonal vaccine which would be delivered on its own at the start of autumn in the northern hemisphere as usual. The first batches of vaccine arrived after the predicted autumn wave of cases had started, slightly earlier than expected, coinciding with children returning to school after their summer break. The people at most risk were prioritised for vaccination: children, pregnant women, people with medical complications and healthcare workers. If a strong immune response was not observed with the vaccine then a second dose would be required so orders were scaled to account for this.
The usual timescale for flu vaccine production, about 6 months, was not possible given the timing of the pandemic outbreak, so various processes were sped up when possible to produce the vaccine before the northern hemisphere winter arrived. Unfortunately, this led to speculation in some parts of the media that the vaccines would therefore not be safe. The WHO and other scientists were quick to reassure that the fast-tracking of the vaccine production process would not compromise the safety or quality control. However, the scaremongering in the media continued, with stories about possible neurological complications, suggestions that the pharmaceutical companies didn’t care if they were safe, use of ‘dangerous’ adjuvants (agents that are added to vaccines to increase the immune response), and the lack of safety testing in people. The swine flu vaccine was created in a similar way to how the seasonal vaccine is created each year, but including the hemagglutinin and neuraminidase from the new strain. The vaccine has been shown to have an excellent safety profile.
In reality, vaccines are not 100% safe. There are very rare cases where individuals have a severe allergic response to vaccination. The benefits of vaccination far outweigh these extremely rare risks, which are continually researched and minimised. Influenza vaccination programs save hundreds of thousands of lives each year around the world. Vaccine uptake was hindered by these sensationalist scare stories, even the quick response of the WHO and other scientific advisers to assure people of the vaccine safety were reported by some parts of the media with scepticism. It does bring up an important factor in communication between scientific experts, the media and the public, and how scientific experts are viewed and the responsibility of media organisations, especially in cases of public health.
It has been estimated that in excess of 160,000 deaths worldwide were associated with the swine flu pandemic. This is slightly less than the number of deaths attributable to seasonal influenza in a typical year, but it could have been a lot worse. Of the very limited number of confirmed human cases of H5N1 ’bird flu’ (~450), the mortality rate is over 50%. If a form of H5N1 that could pass from human to human developed that was equally virulent, then it could be catastrophic.
The world has never had such detailed knowledge of influenza viruses or been as prepared as it currently is for a pandemic. The rapid sequencing of the genes of the H1N1 novel strain, the collaborations between scientists, doctors and government health departments around the world, the sharing of information between such people and the monitoring of the spread and mutations in the viruses have all been important. The access to information in the media and on the internet, and the production of a successful vaccine and use of antiviral drugs in severe cases have been major successes in the response to the pandemic, without which many more lives would have certainly been lost.
In terms of what could be improved, it is often easy to criticise actions with hindsight after the event. Evaluation of the effectiveness of response and what decisions were made is an important part of planning for the future. That the 2009 H1N1 pandemic was not worse was very fortunate, but it doesn’t mean that this will be the case when the next influenza pandemic arrives. The WHO and government health authorities around the world have learnt a lot in terms of which parts of their responses to the pandemic worked well, and which could be improved on.
One aspect of the swine flu pandemic response that is commonly criticised is the early decision to produce a vaccine. Because of the time taken to produce a vaccine, the decision was taken early. However, at that time little was known about the morbidity and mortality of the pandemic virus strain. One key lesson is that comprehensive, reliable serological testing is needed early on in order to make a well-informed decision on how dangerous the strain is likely to be from the numbers of people who have been infected. With greater epidemiological and serological information early on in the pandemic, the scale of vaccine production and antiviral drug orders would probably have been scaled back, reducing the massive cost.
From the standpoint of the post-‘mild’-pandemic phase we are now in, the scepticism of some commentators about the proposed risks in spring 2009 seem fair in an ‘I told you so’ way. But this is to ignore the very real risk that existed. The virus may have been much more virulent, caused more severe disease, or may have mutated into a more dangerous strain before spreading through the world population. Unfortunately, predicting whether a new pandemic will be as mild as the 2009 one, or as deadly as the 1918 outbreak, is currently impossible.
There have been calls for greater transparency, particularly at the WHO, over who made the recommendations of vaccine production and drug stockpiling and their connections to the pharmaceutical companies producing vaccines and drugs. There is no need for media conspiracy stories to run wild, but openness is needed when decisions influence massive public spending. The WHO has taken the need for transparency seriously and addressed concerns that were expressed.
The UK spent approximately £1 billion on stockpiling antiviral drugs and producing vaccines. The stockpiling had been initiated some years previously in case of a H5N1 ‘bird flu’ outbreak. There can be no doubt that several pharmaceutical companies have made a lot of money out of the pandemic, although it should be remembered that no-one else is in a position to produce such vast quantities of vaccines quickly enough. The drugs and vaccines have undoubtedly saved lives and, to put the cost in context, even a massive cost like £1 billion is only about 1% of the NHS annual budget. Government health departments may well be considering how they negotiate with the pharmaceutical companies, for example, it may be in the interests of the health budget to have a more flexible approach to orders, especially when it was found that a single dose of the vaccine was usually sufficient for immunity to develop. Despite the scepticism of vaccine uptake in some countries, it has been estimated that 350 million doses of vaccine were administered and the vaccine was ~95% effective. It is difficult to calculate how many lives have been saved by vaccine uptake, but it could be a huge number. In the future, people are looking at ways to produce vaccines more quickly, using technological advances and better processes, in preparation for the next pandemic.
It is expected that the swine flu pandemic virus will become a seasonal virus, and looks to have displaced the old seasonal H1N1 virus lineage. The new swine flu H1N1 strain is included in the seasonal vaccine for winter 2010-2011.
As the WHO director-general Margaret Chan said in her postpandemic conference statement, ‘this time around we have been aided by pure good luck’. The pandemic virus was much milder than it might have been, the virus did not mutate into a more lethal form during the pandemic, there were very limited cases of Tamiflu resistant viruses and the vaccine was a good match for circulating strains.
The 2009 swine flu has been a significant test for our pandemic preparedness, and it was thankfully mild in its severity. We have learnt a lot, which can only be a good thing, as the next pandemic might not be so mild.