The HRA recently carried out a survey as part of our strategic objective to improve research transparency. The survey aimed to establish some of the current obstacles to transparency, and to identify future opportunities to improve practices.
Why did we run the survey?
We wanted to see what researchers, researcher managers, sponsors and funders understand by the phrase research transparency: what it is and why its important. We also wanted to hear what they think about their own practice in relation to research transparency, and what helps or hinders them.
The survey was not designed to check compliance with research transparency requirements or best practice. We are using the findings to inform our policy work and our communications to the research community. Most of the questions asked for a free-text response, so the results give us an impression rather than any firm conclusions.
Who took part in the survey?
We ran the survey for a period of six weeks from August 2018, promoting it on our website, on social media and in our monthly newsletter. We also sought support from stakeholders, asking them to pass on the survey to maximise its reach.
In total, 266 people responded to the survey. Of those that told us about their role in research, more than half (79/139) said they are a researcher in the NHS or in a university, as the graph shows.
Understanding and awareness of research transparency
Our first, open question looked for ‘top-of-the-head’ awareness and understanding of the term ‘research transparency’. When analysing the answers to this question, we looked for unprompted mentions of one or more of the ‘four pillars’ of research transparency. They are:
- registering research
- publishing and disseminating findings and conclusions
- giving access to the data and tissue used in the research
- giving information at the end of research to participants
We found that relatively few respondents mentioned the four pillars of research transparency, and none of the respondents mentioned all four aspects of research transparency.
Whilst publication was most often mentioned, fewer mentioned registration, and fewer still making results public by posting summary results in a register or publishing in an open access journal.
Only one respondent thought that research transparency means giving information to the people who took part in the study.
Over half of respondents (142/266) gave a different but specific definition of research transparency, including:
- Being open about methods, sources of funding and conflicts of interest
- Providing information to participants (for example through a participant information sheet), especially the information specified by the General Data Protection Regulation about the processing of personal data
The response to this question suggests that there is a general awareness of research transparency amongst researchers, research managers, sponsors and funders. However, most instantly think of broader ideas like openness about funding or research methods and clear information about the study for participants. These are important features of research, but they weren’t the aspects of transparency that we were expecting to see.
We will use these findings to inform how we communicate with the research community about transparency.
How well do researchers think they are doing?We went on to ask respondents about their current practice in relation to research transparency:
Practice in relation to publication was better than in relation to registration.
relation to making findings available to participants was more variable, with
around half (126/233) saying that they always or mostly do this and a quarter
(61/233) saying that they didn’t know or it didn’t apply to them.
Finally, we asked about current practice in relation to making data or tissue accessible to other researchers:
The answers to this
question suggest that, although respondents’ unprompted understanding of
research transparency was relatively poor, their actual practice is better.
Only a quarter understood registration to be a feature of transparency, yet
more than half actually do it all or most of the time. Similarly, around a half
saw making results public part of transparency, yet two thirds actually do it.
Finally, only one respondent understood transparency to mean telling participants about the findings, whilst half of them actually do it. Having said that, it is worrying to see that, at best, only two thirds of respondents are meeting expected practice in research transparency. We will be addressing this through our policy work in this area.
What helps and what hinders?
Finally, we asked respondents to tell us about the things that help them meet their transparency duties – and those that get in the way. The responses to this question were wide ranging and it is not possible to draw any reliable conclusions. However, not knowing where to register was the most commonly cited obstacle to making information about a study publicly available before it starts.