Crop protection: managing risk, not avoiding it
Extract of a speech by Sir Mark Walport to the Crop Protection Association on the current science around neonicotinoid insecticides.
On 14 May 2015 Sir Mark Walport gave the keynote address at the Crop Protection Association (CPA) annual convention. Sir Mark discussed the themes of his annual report ‘Innovation: Managing risk, not avoiding it’ and how they apply to the work of CPA and its members. An extract of his talk, which focussed on the current science around insecticides, particularly neonicotinoids, is available below.
It would be odd at a meeting of the Crop Protection Association not to talk a little about the issue of insecticides - and specifically about one class of insecticides, the neonicotinoids. And it is important to put these into the broader context. There are 2 broad reasons for using insecticides, the first is in the protection of crops against insect pests, the second is to protect humans and other species against insect borne diseases, such as malaria, dengue and sleeping sickness. And in both cases, there is a risk/benefit assessment, balancing the harm to humans and other animals from disease or loss of crops – against the harm to the environment, including bystander insect species that are important parts of the planetary ecosystem.
As Government Chief Scientist it is important, together with my colleagues Ian Boyd at Defra and Chris Whitty at DfID, that we provide the best scientific advice to policymakers - and that includes communicating uncertainty and updating our advice as new and better evidence emerges. One of the lessons from my own scientific area, which is medicine, is that proper ‘clinical trials’ need to be undertaken, conducted in the ‘real world’, in this case the field, rather than the laboratory. These trials must be adequately statistically powered to deliver robust results. The second is that it is rarely possible to make policy decisions based on single studies – and one of the great contributions to medicine has been the meta-analysis, the Cochrane Review, conducted objectively and reviewing all of the evidence that is germane to a particular area of activity. That is why the work of the Intergovernmental Panel on Climate Change (IPCC) has been so important in the area of climate science. And I will come back to this with respect to neonicotinoid insecticides in a moment.
So the key questions in respect of neonicotinoid insecticides are: firstly, what is their hazard? And that is an easy question – they are highly toxic to many species of insects; secondly, what is the risk of their application in agriculture? That is a much harder question – and this depends on the nature of the exposure. So it is immediately important to distinguish between the application in the form of treated seeds or as foliar spray. And this is where the ‘clinical trial’ is important – we know from the application of neonicotinoids in the laboratory that they are harmful to different species of insects – but this is very different from knowing what happens when insects are exposed ‘in the field’ to neonicotinoids applied under agricultural conditions.
So in this context the recent Swedish trial, reported in Nature in April of this year is an important contribution to the evidence base - this was looking at rape seeds treated with both a neonicotinoid clothianidin, and a pyrethroid. The scientific team found evidence for a reduction in wild bee density, solitary bee nesting and bumblebee colony growth and reproduction under field conditions. Interestingly honeybee colonies did not differ in strength between treated and untreated fields - which taken with other evidence indicates that honeybees are more resistant to neonicotinoids than other species of bee. A second trial, similar to this in design is also being conducted in the UK by the Centre for Ecology and Hydrology and I look forward to seeing the results of this study.
The authors of the Swedish study also made the following important statement:
We question whether prevailing risk assessment standards, where predominantly short-term and lethal effects are assessed on model species under laboratory conditions can be used to predict real-world consequences of pesticide use for populations, communities and ecosystems.
I strongly agree with that statement and have no doubt that the world of plant and animal science would benefit strongly from assimilating the lessons from the clinical trial, with all of the strengths and weaknesses of clinical trials, accompanied by rigorous meta-analysis.
So on the subject of meta-analysis I would like to acknowledge the work of Professors Charles Godfray and Angela McLean, who last year published with a group of distinguished colleagues in the Proceedings of the Royal Society B ‘A restatement of the natural science evidence base concerning neonicotinoid insecticides and insect pollinators’. Charles and Angela have agreed to update that paper to include papers that have been published subsequently.
But there is a further aspect to the clinical trial that is important - clinical trials are about efficacy at least as much as about toxicity. There are 2 approaches to the application of pesticides – as prophylaxis and as treatment - and the clinical trial approach is needed for each of these. In the context of this as well, I look forward to the publication of updated meta-analysis from Charles and Angela.
My final points on insecticides are firstly to beware of unreasonable extrapolation of evidence, for example from seed treatment as prophylaxis to foliar treatment for infestation - and secondly to examine very carefully what are the consequences of cessation in use of one class of insecticide - will this result in the use of other compounds instead and what will be the effects? And - one of the greatest scientific contributions to the prevention of malaria has been the introduction of insecticide-treated bednets and household spraying with insecticides. Dengue control depends on the application of insecticides. Throughout the world there is a spread in resistance to the insecticides that have been key in the control of these and other insect born diseases. And there are too few new insecticides in the pipeline, because the market incentives to develop these are so weak.
So my message to this meeting on the subject of insecticides is that the way forward is by the rigorous application of the scientific approach to gathering evidence - and that scientists should concentrate on this and not stray into advocacy.