Around half the antiviral drug Tamiflu that was prescribed during the 2009 H1N1 swine flu pandemic was never used, researchers have discovered.
The scientists looked at how much of the drug was taken within two populations, 208,000 people in Oxford and 6,000 people in nearby Benson, by analysing the sewage they produced over 24 hours. Like most drugs, all the Tamiflu people consume ends up flushed down the toilet, so analysing its presence in wastewater is a fast, cheap way of telling how much has been taken.
An estimated 600,000 courses of Tamiflu went unused over the course of the pandemic, at a cost of around £7.8 million to the UK taxpayer. More seriously, this suggests that the government's plans to respond to flu pandemics need to be improved. The 2009 swine flu turned out not to be as deadly as had been feared, but if it had been more serious, would people still have failed to take Tamiflu they'd been prescribed? If so, they'd have been endangering both themselves and others.
'The UK government's plan ended up working well, but in a severe epidemic or pandemic people who get Tamiflu and then don't use it are taking it from those who really need it,' says Dr Andrew Singer, a chemical ecologist at NERC's Centre for Ecology & Hydrology and member of the UK government's Scientific Pandemic Influenza Advisory Committee, who led the research. He argues that much greater cooperation between medical and environmental researchers is needed to let us understand these problems and respond more effectively.
The study gives valuable insight into how people behave during disease outbreaks, but it suggests much more research is necessary. Singer suggests we particularly need to understand why people get hold of medicine and then don't take it. Different groups might do so for different reasons; previous studies have shown that they certainly vary widely in their willingness to follow doctors' instructions. If we knew more about the subject, the government could do more to persuade people to take their medicine, perhaps with better public-awareness campaigns and targeted advertising.
He notes that the government's plans for dealing with pandemic flu generally assume people will take the antiviral drugs that are prescribed to them. If they don't, things could go badly wrong. For instance, many people with flu will develop secondary infections like pneumonia, which need hospital treatment. Emergency plans exist to handle this, but they take into account the fact that Tamiflu cuts the rate of secondary infection by around half. If people don't take their Tamiflu, there will be more secondary infections than expected, and hospitals won't be able to cope with the influx of patients.
'There are a lot of hidden effects that we don't understand very well,' Singer says. 'It's my feeling that many people are quite reluctant to take medicine but like to know it's available should it be absolutely necessary. And here is where the subjectivity comes in – what makes one person pause while another adheres to the doctor's orders? There needs to be a better understanding of people's behaviour under these circumstances so that public health messages can be as effective as possible and drug stockpiles are not wasted.'
We don't yet have a fast, reliable test for flu, so it's not possible to ration Tamiflu to people who really have it. Instead, Singer says we need to understand why they fail to take drugs, and use this information to persuade them to change their ways.
He argues it would help if the government shared its very detailed records of how much Tamiflu was prescribed in each area with environmental researchers. 'It would be hugely beneficial if the health sector would work more closely with environmental scientists,' he says. 'Existing surveys on Tamiflu compliance often only capture a couple of hundred people and are limited to certain demographics; our study captures more than 200,000, and covers the whole population at a fraction of the cost.' The benefits aren't confined to antivirals – similar methods could be used for nearly all drugs.
The study was published in PLOS One.
Singer, A. et al. (2013) Compliance to Oseltamivir among Two Populations in Oxfordshire, United Kingdom Affected by Influenza A(H1N1)pdm09, November 2009 - A Waste Water Epidemiology Study. PLoS ONE 8(4): e60221. doi:10.1371/journal.pone.0060221