Turnbull’s stated aim of an “agile” society brings to mind a trapeze artist. But while all eyes are fixed on the acrobat soaring at death-defying heights, the safety net lies below, ensuring no one breaks their neck.
To realise the promise of innovation, we need to equally embrace the need for comprehensive and adept measures to assess and manage the risks of industry-researcher collaboration.
The risks of collaboration
The innovation report nowhere mentions the risks of collaborating with industry: risks of bias and conflict of interest.
These can divert the entire research agenda so that it privileges certain interests, distort the true effects of research and break the public’s trust in science.
Commercial and public interests are not always aligned. This is particularly the case in industries whose products may be harmful to health such as food, drugs, medical devices and chemicals.
The tobacco industry is a classic example of research partnerships that were used to undermine tobacco control policy and to generate scientific doubt about the harms of smoking.
In the medical industry, relationships with clinicians have made marketing a routine part of health care. This drives up costs and puts patients at risk through the promotion of drugs and devices with limited track records for safety.
Industry sponsorship and risk of bias
The quality, credibility and cutting-edge nature of research depends on managing risks of bias.
Risks of bias are systematic errors in research that may lead to an underestimate or overestimate of the true result. Risks of bias are different from what we commonly understand as bias in the prejudicial sense – they are independent of a researcher’s beliefs or ideologies.
Industry sponsorship of research is a known risk of bias. Studies show that when drug or device manufacturers sponsor research to test their product’s efficacy or safety, the results and conclusions are more likely to be favourable than for studies with other sources of funding. This is true even when the studies’ methods and quality are identical.
This means that industry can fund and conduct high-quality research, but that the results may still be distorted, exaggerating either the benefits or minimising the harms.
The innovation report proposes rewarding researchers who engage with industry, but will measure this in terms of research income from industry. This kind of funding arrangement poses a high risk of bias to research endeavours.
If we are to be sure that an innovative medical device, software application, industrial chemical or new agricultural practice is effective and safe, those who stand to gain financially from its commercialisation and those who evaluate its benefits and harms need to be independent.
Bias can enter the research process at any stage, but it also has an amplifying effect – bias that shapes the kinds of research question that even get asked (or funded) is magnified at the level of public policy.
For example, at the Charles Perkins Centre, researchers are examining the association between funding source and the focus of nutrition research. If the research agenda is biased, down the line, dietary guidelines may in fact misguide the public.
Innovation and conflicts of interest
Researcher conflicts of interest not only threaten research integrity, but public trust in the credibility and legitimacy of science itself. It should be that researchers can proudly collaborate with industry for public benefit.
But researcher conflicts of interest – including the wrong kinds of partnerships, failure to disclose personal financial ties and failure to maintain independence – can put an entire field’s work into question.
For example, the recent disclosure of the millions of dollars that Coca-Cola paid researchers calls into question public health research that focuses on a lack of physical activity as the main culprit for obesity, while ignoring sugar-sweetened beverages.
Increasing the impact of Australia’s world-class research in communities at home and overseas is a goal researchers, industry and governments share.
In partnering, it is essential that researchers maintain their independence and that innovations are evaluated objectively.
This means that freedom to publish, disclosure of financial ties and management of conflicts of interest need to be part of the innovation strategy going forward.