Since the ANZAC day weekend when many Australians first learned about a novel strain of influenza circulating in Mexico and the United States, there have been different views among health professionals, politicians and the population about how we should react to the influenza pandemic of 2009, still widely referred to as swine flu.
The sorts of questions that people were asking in the early days, and are continuing to ask, were the very obvious things we wanted to know immediately. How serious is swine flu in relation to other seasonal strains of influenza and what are the best ways to manage the pandemic? One general practitioner, writing from Victoria, wondered if we got away with our approach to the pandemic because the virus was relatively mild. As the epidemic moderates, we are in a position to explore some of these issues.
Unlike countries in the northern hemisphere, Australia experienced the swine flu outbreak when seasonal influenza circulation was expected. Victoria was the first Australian state to report a significant number of cases and the first Australian state to record the epidemic peak. This occurred around the end of June.
The peak was similar to peaks recorded for seasonal influenza in 2003 and 2007, but the 2009 peak may have been inflated by increased presentation to general practitioners. Worried about swine flu, people with an illness that they thought may have been influenza presented to their general practitioner in 2009 when they might have stayed home in any other year.
If this were the case, the swine flu peak comparison, based on the proportion of people presenting to their GP with an influenza-like illness, may not be the most informative comparison.
Another way to look at influenza data over years is to calculate the number of tests that are positive for influenza as a proportion of all tests requested for influenza. Surprisingly, this simple proportion appears to allow comparison of influenza seasons between years and between Australian states. The method corrects for increased testing, which might be associated with higher numbers of influenza diagnoses, but not necessarily higher levels of influenza circulation in the community.
In recent years the proportion positive confirms what was known from other surveillance systems, that 2007 was a year with high levels of circulating influenza viruses. But, for Victoria, the proportion of flu tests positive in 2009, when swine flu was far and away the dominant circulating strain, was about the same as the proportions in 2004 and 2006, years known to be characterised by relatively low influenza activity.
What we can now say is that the first experience of a swine flu season in Australia, the Victorian experience, suggests that influenza circulation in the community was at most like a season characterised by moderate seasonal activity, but it is also possible that flu activity was lower than this. And there were hints about low level swine flu activity from the northern hemisphere. Google Flu Trends, which has accurately charted the Victorian influenza season this year, has charted a mild flu season in Mexico 2008-9, during which time swine flu emerged.
The conclusion about relatively low-level activity has recently been confirmed by a group of scientists who have modelled the swine flu effect in Mexico. From the outset numerous commentators, in Australia and abroad, have made the point that swine flu was a relatively mild disease, a fact also acknowledged by the World Health Organization.
However swine flu causes a wide spectrum of outcomes from infection without disease (asymptomatic infection) to serious illness and death. We can’t count people who have no symptoms because we can’t identify them. For seasonal flu, the proportion with asymptomatic infection is around a quarter to a third. Modelling suggests it may be even higher for swine flu. Because we can’t identify people with asymptomatic infection and because many people with swine flu were never tested, representatives from the World Health Organization now admit we have no idea how many people have been — or will be — infected with swine flu.
They suggest the number may eventually rise to one third of the world’s population. This might be around the proportion infected by seasonal flu in a bad season, but we can’t be sure of that, either.
Between the extremes of asymptomatic infection and death are the people with swine flu symptoms, ranging from a bad cold to symptoms severe enough to warrant hospital admission. Some hospitalised patients need to be cared for in Intensive Care Units. This is where swine flu infection has differentiated itself from seasonal flu. We expect people with underlying respiratory and heart disease to suffer worse outcomes from influenza infection. We have seen this in patients with confirmed swine flu infections, but with seasonal flu infections we do not see the high rate of serious disease amongst pregnant women and people who are very overweight that we have seen with swine flu.
This reminds us that, even if evidence now accumulating suggests swine flu may be no worse — or even not as bad — as some strains of seasonal flu, we must not be relaxed about a disease which has the potential to cause serious disease.
People also die from swine flu, but probably not as many as the numbers thought to die from seasonal flu. In Victoria, as the swine flu epidemic appears to wane, the number of deaths recorded as due to swine flu in mid-August was 21. We don’t know the numbers who die from seasonal flu each year because we don’t test for seasonal flu with the same enthusiasm that we have tested for swine flu. However modellers have given us some idea about this. For Australia it is estimated that between one and three thousand people die from seasonal flu each year.
In Victoria this would equate to 250-750 people, an order of magnitude higher than the number of swine flu deaths to date. Unfortunately there will be more deaths due to swine flu but it seems very unlikely the tally will reach even the lower limit estimated from the models.
Nonetheless swine flu, like seasonal flu, causes severe illness and death and, like seasonal flu, it should be preventable by vaccination. Vaccine manufacturers in Australia and abroad are well advanced with a vaccine aimed to protect against swine flu. Since they are manufactured in the same way, we should expect these vaccines will be as safe and as effective as seasonal vaccines. Swine flu is a novel virus, so we are not yet sure whether adults will need one or two doses to provide protection.
As a first priority, authorities here and abroad are concentrating on vaccination of health care workers and those most at risk of an adverse outcome from swine flu infection, based on the best available information.
However the best available information does not provide a complete picture of what is going on. We have seen there are many things we don’t know about swine flu. How many people have been infected? How many more are likely to be infected? How many are likely to die? But, hang on, these are the same things we don’t know about seasonal flu. In previous pandemics, the pandemic virus displaced the dominant flu virus circulating in past years and the pandemic virus became the dominant virus in subsequent years.
We don’t know how this occurred or if it is going to occur this time. Pandemic waves have been described for the three pandemics of the twentieth century but commentators have recently suggested there is no good evidence for waves in the two most recent and milder pandemics.
So, waves might not be a universal phenomenon of pandemics and, even when they do occur, we are not sure why. In the 1918-19 pandemic, the mother of all pandemics, was the wave phenomenon due to relaxing social distancing measures, was it something about the virus, was it a result of different attack rates in different ages with subsequent changes in immunity or is there another explanation?
Surveillance, control, treatment and prevention of influenza is a multi-billion dollar global industry spanning the public and private sectors with many de facto public-private partnerships. However, this impressive and important edifice is built on surprisingly shaky foundations. In no country where influenza treatment and prevention is provided at least in part from the public purse, do we really understand the annual burden of disease proven — not modelled — to be due to influenza.
In all the cost-effectiveness studies supporting the use of influenza anti-viral medication and influenza vaccination, the burden of illness data used in the analyses are derived from modelling, relatively small trials or, quite often, values someone else used in a similar analysis.
Unfortunately this is not the only area of surprisingly poor knowledge about how we manage influenza. Influenza vaccine, not the best of the vaccines that are publicly funded but certainly one of the cheapest, is generally accepted as being effective in preventing symptomatic infection in about 70% of healthy young adults. But healthy young adults are not the people we target for vaccination. We target people aged 65 years and over.
Agreement on the effectiveness of the vaccine in preventing infection and death in this age group is another matter, however. For more than a decade it was thought that influenza vaccine could prevent around 50% of deaths due to any cause in people aged 65 years and above.
However a few years ago researchers in the US showed there was no decrease in modelled excess deaths due to influenza in this age group over a 20-year period when the influenza vaccine coverage increased from 15-20% to 65%. This counter intuitive finding lead other researchers to suggest that flaws in the studies showing 50% protection against death accounted for the 50% finding. But the researchers who suggested that 50% was a plausible estimate re-analysed their work to confirm its validity. The issue remains unresolved.
As the pandemic continues to wind down in Victoria, I am wondering why can’t we learn some lessons from the whole thing? We spend billions on anti-virals and vaccines. For instance, the Australian government has just allocated another $44 million for influenza vaccine for people of any age at increased risk of an adverse outcome from influenza infection.
This is probably a great idea but we do not have the data to know for sure. Although time consuming and expensive to collect, why can’t we assemble quality data on the basic epidemiology of laboratory confirmed influenza over a few influenza seasons? There is no study design barrier here.
Why can’t we conduct randomised controlled trials on influenza vaccine effectiveness? Again no problem with study design, although researchers might have trouble in the ethics committee. If we had this information we could make truly rational decisions about the cost-effectiveness of various interventions aimed at controlling influenza and we could make sensible comparisons between seasonal and pandemic flu.