Probity can be defined as uprightness, moral correctness and honesty. In research terms the meaning remains the same and in this blog I am really talking about integrity in the conduct of research.
When we undertake medical research we should do it with a fully open mind and with a view that we will frame a question and whether the answer is what we expect or not, the result itself has value and is interesting.
So that is the theory, the reality however, can be quite different.
I have been thinking about what is it that drives people to be ‘economical with the truth’ or plain fraudulent in medical research? I think there are a number of answers to this question and some of them relate to the university, health service and publishing cultures. Despite what I have said about the outcome of research being interesting whatever it shows, the reality is that journals are less forthcoming about publishing well designed studies that show negative results compared with studies that show positive results. The same can be said of grant funding agencies and although a negative outcome may close one hypothesis and open a new one, it is often just seen as a negative result.
On the other hand Universities and Health providers usually measure value in academic attainment on the basis of peer reviewed publications and their impact factors and grant income. The interdependency of this assessment of value and research outcomes is clearly evident.
So we have created a research culture that favours positive results in terms of reward (grants and publications) and also favours positive results in terms of personal development and institutional recognition. It is small wonder then that some individuals feel pressured into crossing the line of research probity and integrity. This of course does not make it right but it may go some way to explaining why it happens. I have worked in labs where consistent failure to obtain breakthrough grants or publications meant an individual would have to leave (employment) and so for many the stakes are very high.
Academic researchers frequently invest huge time and thought in creating hypotheses and for many the academic paradigm makes it very difficult to discard these ideas. I feel that academic researchers sometimes go to extraordinary lengths to prove a point or demonstrate some positive result. This is partly due to the requirement to meet the markers of academic attainment as well as to provide a report for the grant funding agency which is sufficiently impressive that it counts for the researcher rather than against. The reality is that negative outcomes for well designed studies are very important because of the money and time that they can save by steering other researchers towards potentially more fruitful pastures.
Contrast this with industry. In industry led research, for example drug development, compounds which don’t show promise get instantly ditched. There is not the same emotional investment in hypotheses. Research of this type may be in many ways said to be more efficient. That is not to say that it is better, because often industry-led research lacks the depth of investigation and understanding that is attained from academic research. Certainly I believe that research probity is often better in industry than in academia partly because the culture of research is also different. In addition the consequences of research fraud for a company can be fatal and so quality control and governance may be more rigorous. But lets be clear, financial drivers and career progression can still drive research fraud in industry.
How can we alter drivers to commit research fraud?
What needs to happen is a change in culture. One suggestion to achieve this would be to move to a system where research is measured on the basis of quality of design and its execution rather than simply on outcome. Readers will be familiar with the huge numbers of studies that show positive outcomes but with little statistical power and incur a risk of false positive reporting. It is a sad enditement of the times that these articles often get published in higher impact journals than the well designed and adequately powered follow up study which shows that the previous positive reports represented a false dawn.
Such a culture change would require strong leadership from journal editors and from the designers of national research assessment schemes such as the UK RAE or REF. Whether these individuals can rise to the challenge remains to be seen. There is some hope of culture change and the new REF scheme for the first time will take into account the clinical impact of research and this may offer some hope to individuals for a more balanced appraisal of the value of research that has both positive and negative outcomes particularly where this shapes the development of clinical practice or research direction.
The other aspect of culture that needs to be changed is to give research methodology a higher priority in the medical curriculum. Currently I believe that in many places teaching and training in experimental design is often done poorly for medical researchers. In the same way critical appraisal is not afforded much importance. We need to create the norm where doctors are fully equipped to undertake good research and to understand research undertaken by their peers. The argument exists that the already crowded medical curriculum cannot accommodate more content. The challenge is therefore to find a way to improve our teaching of doctors about how to conduct and understand research.
Coming soon plagiarism in medical research
© 2012 SJ Wigmore