select search filters
briefings
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

COVID-19: Why we need to hear the evidence directly from the scientists

By Fiona Fox

This blog contains the thoughts of the author rather than representing the work or policy of the Science Media Centre.

 

I recently asked a SAGE member if he was able to enjoy anything about his role. Quick as a flash, he said “about two hours (enjoyment) to half an hour (hell)”. On reflection I think I might be about the same. Like everyone I hate the virus with my whole being but there is so much to celebrate about the response. Every day I see the most amazingly brilliant scientists take a break from punishing research schedules to speak to journalists. And fantastic science press officers who guide and support their scientists through multiple media encounters while being pulled in a hundred directions. Or I watch as the science and health journalists report clearly and accurately day in day out in the face of bewildering uncertainty and complexity. Those 2 hours can be properly joyful.

But my god the half hour can be bad. And in my case, it’s usually when scientists are prevented from reporting their data to the media by government communications staff.

Like everyone involved in responding to COVID we should be asking ourselves what if anything could be done better. After all everyone acknowledges that the public understanding of response to this virus is not a ‘nice to have’ but a critical factor to beating it.

 

So here’s the multiple choice question;

What do you think is the best way to get new data and science on COVID-19 to the wider public and policy makers?

  • 1) Via the scientists who have gathered that evidence presenting it to science and health journalists at a press briefing where they can answer questions and ensure the journalists understand the complexities such that it is reported accurately.
  • 2) By ministerial announcement on one flagship radio or TV programme, something like the Andrew Marr show or Today programme.
  • 3) By government press release sent out with no opportunity for journalists to hear from scientists doing the work.

 

Yes, these are rhetorical questions, but they are no less true for being loaded. Too often the scientists who are doing the research and gathering the data are being robbed of the opportunity to present their science to the public in the best way. I think it matters and this blog is an appeal to everyone involved in this part of our pandemic response to reflect on whether we could be doing this better

Tempting as it is, I can’t give you chapter and verse on each of these situations because it would upset too many people and the SMC needs to maintain the trust of our friends in science.  

But, let me give you some sense of what happens at our end.  And I should stress here that our end is the only vantage point I have. Those reading this at the government or university end may have a very different interpretation and I would be delighted to discuss this with them or publish their comments on this blog.  

I guess it all starts with a part of government either asking an academic group to do some important work on COVID – whether that’s clinical research on a particular aspect of the virus or the various modelling or surveillance studies. This commission might come from DHSC, PHE or BEIS. The scientists who do the work are often primarily based in universities and their press officers are university press officers. But because the studies are in collaboration with government departments or agencies, the university press officers need to agree a comms plan for announcing any findings with the government communications people. They in turn must get sign off for the plan from other comms officers and special advisors across government departments, including perhaps Cabinet office, Number 10 and so on.

The SMC comes into this chain because we are in touch with the Principal Investigators (PIs) and scientists leading these studies and their press teams, and we are running almost all the press conferences on the science around COVID-19. We get to know when each group is due to publish updated data or announce the start of or early results from trials. Many are publishing data regularly and we are on hand each time to run a press conference for the science journalists. To the huge credit of the scientists doing this work, most of whom are working 12 hour days and 7 day weeks, they are enthusiastic to take part in these briefings. This is in part because they get to speak to all the science journalists in one go rather than dealing with multiple interview bids, in part because they want to see the data reported to the public in the most accurate way, and hopping on a zoom briefing to answering questions from well informed science and health journalists is the best way to do that. I wish you could listen in to some of these briefings where scientists and science journalists grapple with the best, most accurate, most accessible ways to report complex data. It is something the UK should be proud of. These briefings are definitely up there in my 2 hours of feeling good.

I should emphasise that this is not happening all the time. After all we have run 74 briefings on the science of COVID-19 and many of those were with researchers who are government funded. When, for example, UKRI and NIHR are allowed to be in the driving seat, things go much better. But the fact that it can work so well with some but not others spurs me to push for better. And it’s happening enough to make life pretty miserable for groups of scientists and press officers who deserve better.

It happens in 3 different ways. Comms people in government send word to the university press teams and/or academics that they don’t want a press conference for a particular set of findings. Or they explain that permissions have not been received from other bits of government comms including number 10, so nothing can be scheduled or advertised to journalists. Another way is that government comms on the science goes out earlier than planned via a ministerial announcement on something like the Marr Show or at one of the Downing Street press conferences. More than once I have been called by journalists on the Today Programme at 6am desperate to reach the scientist doing the study after a DHSC press release has just landed. On some occasions when I contact the relevant university press office, they haven’t even seen the release or been told that the announcement was being brought forward.

 

Do you see why this is frustrating? And do you see why it matters?

 

So first why do government communications staff block scientists communicating research and data. Well I don’t know, and they don’t tell me, but here are my guesses:

  • 1) Because they can – because they fund or commission these academics and therefore it’s legitimate for them to take a lead in deciding how those results get into the public domain.
  • 2) Because they are spinning a hundred plates at once and trying to co-ordinate these announcements across key government departments, and it’s sometimes just impossible to get all the relevant sign offs in time.
  • 3) Because departmental comms officers are essentially political with a small p. Their job is fundamentally different to that of a science press officer working in a university and they need to focus on how any science announcements fit in with government messaging on COVID-19.
  • 4) Because ministers want good news announcements which are in short supply in this crisis, so news on new trials or treatment breakthroughs can be a nice Sunday story, or a positive announcement for the Downing Street press briefings.
  • 5) Because government press officers are trying to exert some control on overall messaging. If Boris is running a press conference to announce stricter measures or further lifting of measures, they prefer to be in control of the timing of any data released around that.
  • 6) Because comms staff at PHE, DHSC are in permanent firefighting mode and don’t want something said by one scientist at an SMC press briefing to start another fire.  A short press statement or a ministerial announcement is more under their control and less risky.

 

There is no great scandal here. It’s not the case that science is being delayed or suppressed or misrepresented. All this data is published in a timely manner. And in fact, on several occasions when briefings have been blocked, the SMC has gone ahead with a press conference anyway, even when the story is already out there. Nor am I amongst those who are angry at the lack of openness during this crisis. I am on record as celebrating greater openness from scientists on COVID than in past crises. When we have got the right permissions in place to allow SMC press briefings, we have been pleasantly surprised by the lack of meddling from government comms officers.

But the fact that there is no scandal doesn’t mean there is nothing at stake. There is a better way of doing this and when the stakes are this high, we should all try to do things better.

 

So here is my heartfelt plea to everyone involved:

 

To government press officers in PHE/DH/BEIS/Number 10 etc

Treat science announcements differently to standard policy announcements. Have regard to the fact that the very best way to get science and data into the public domain is through the scientists explaining it to science journalists who then report it accurately to the wider public and policy makers. This is good news for government as well as the public. If the science is misreported, you will have to deal with the fall out. Separate government comms from science comms such that the university or research institute press officers always lead on the research comms. These press officers are smart and responsible and will liaise with you and act responsibly. But please understand that it is not trivial to prevent scientists presenting their data to the media and any sticks should be bent very firmly in the direction of not doing so. Remember that allowing scientists to communicate independently would almost certainly be good for government. Polls consistently show that independent scientists are trusted much more by the public than governments.

 

To the University/research institute press officers

Ask yourself why academics at your institution should be told what to do by someone in government communications when everyone else believes it’s in the public interest to get it into the media in the best way possible. Push Back. In our experience government comms officers tend to lead on communications in part because scientists and science press officers let them. You could start by not asking permission. It’s soul destroying to see some of the emails exchanged between science comms officers and their peers in government where permissions are being sought at every stage. Some of the briefings don’t happen in time simply because university press officers wait too long for permission that never arrives. Also ask for the reason. Of course, there may be good reasons why government don’t want a science press briefing or want it at a certain time. None of us want to clash with a SAGE briefing or a Downing Street press conference. But getting no explanation or being told it is because Number 10 has not signed it off is not a good enough reason to deny journalists and the public access to new scientific data presented in the best way.

 

To the scientists

I understand that you have relationships with scientists and officials in government, and don’t want to damage those in a way that could ultimately undermine your science. But you need to make a stand here. Communicating your science in an open and unrestricted way is important to the scientific process. You would not accept political interference in your science, and you should resist it when it comes to communicating it. The scientific community should resist becoming an arm of government communications, and I’m convinced that when scientists take a stand the principles of independent communication will start to become the norm. Also get help from your university.  I understand that your Research Office may have agreed the contract with government departments and may have paid little attention to the clauses on communication.  Ask them to pay attention to these clauses.  When I warned the Lancet and University of Oxford that I thought this might be about to happen to some data on vaccines due to be published in the journal, the Lancet press office swung into action, proactively issuing warnings to everyone involved, including all government partners, that it would not be ok for anything to come out of government before the press briefing with the scientists. It was impressive to watch, and it worked.

 

So how can we improve things? Well we can start by more of us acknowledging that this matters and not accepting it with a resigned refrain that ‘government will be government’. I know lots of things are going ok and we can’t fix everything. But I am not making any of this up and if anyone reading it is able to help improve things then please consider doing so.

One of the few successes I have had in this space was to halt the growing encroachment of purdah rules into academic science, where rules designed to stop departmental civil servants from making policy announcements in the run up to an election started to be imposed on scientists in universities and research council units. What was most striking about that campaign was the realisation that no-one in government or science had ever thought it was a good idea. But neither had anyone thought it was their job to stop it. We need The Great and Good of science including Vice Chancellors, the leadership of UKRI, the Chief Medical Officer and Chief Scientific Advisor, and others to add the independent communications of COVID-19 research to the list of issues on their radar.

There are some principles already in place which could easily be adapted to Covid-19 research which state that independent scientists doing work for government should be free to publish and present their research independently.

Another useful model we could look to emulate is the Code of Practice for Statistics which applies the principles of independence to all government bodies who publish statistics, with sanctions for those who don’t comply. The reason I really like this Code is because it enshrines an independence of statistical function inside central government. My feeling is that we have scientists who want to announce their data independently but who have no line of defence when Government tells them they can’t. Could we have a similar Code of Practise for communicating research commissioned by government? I would happily volunteer to help lead any such initiative.

If COVID-19 teaches us anything, it is surely that one thing that will make a big difference to the course of the pandemic is the public’s response to it. Ensuring that they hear the evidence from the scientists who understand it best and the journalists most likely to report it best is not a side show. It’s essential. 

3 Responses to COVID-19: Why we need to hear the evidence directly from the scientists

  1. Sir Dave King says:
    Fiona, I fully support what you have to say, but I am surprised that you managed to produce this without reference to Independent SAGE and the reasons for setting it up. The scientists on the group have been in the media virtually every day over the past 4 months.. …

    Fiona, I fully support what you have to say, but I am surprised that you managed to produce this without reference to Independent SAGE and the reasons for setting it up. The scientists on the group have been in the media virtually every day over the past 4 months.. There is a real hunger for what they have to say, as top experts in the field, precisely because the Government since 2010 has been placing the SAGE teams and other Science Advisors under their command and control. Please do understand that the crucial importance of openness, honesty and transparency in the Government’s Science Advisory system, as strongly recommended in the Phillips Commission Report 2001 into the BSE livestock epidemic, was the reason for establishing IndieSAGE.

    read more
  2. Robert West says:
    As we all know, restrictions on presentation and discussion of findings are not new, and in fact are the norm for research undertaken under contract to government or commercial organisations. The journal, Addiction, requires authors to state whether they were under any such strictures because even when results are finally …

    As we all know, restrictions on presentation and discussion of findings are not new, and in fact are the norm for research undertaken under contract to government or commercial organisations. The journal, Addiction, requires authors to state whether they were under any such strictures because even when results are finally published, the possibility that they could have been censored biases the research literature.

    While it is easy to see why government may be nervous about scientists presenting and discussing findings unfettered, allowing us to do so would set a new standard and I think would paradoxically free government from having to worry about what the scientists might say – because they could be more obviously distanced from it if they wished to be.

    Scientists must take their responsibilities regarding dissemination of findings even more seriously without this kind of brake. We have to check, double check and get others to treble check the findings because if we get it wrong it will be on our heads.

    read more
  3. Tom King says:
    The code of practice has three statutory pillars for producers of official statistics: trustworthiness, quality, value. But the Office for Statistics Regulation (OSR) already supports other producers to voluntarily adopt the code. And government can apply the principles of the code voluntarily to outputs that are not official statistics: https://code.statisticsauthority.gov.uk/voluntary-application/ As a …

    The code of practice has three statutory pillars for producers of official statistics: trustworthiness, quality, value. But the Office for Statistics Regulation (OSR) already supports other producers to voluntarily adopt the code. And government can apply the principles of the code voluntarily to outputs that are not official statistics:
    https://code.statisticsauthority.gov.uk/voluntary-application/

    As a first step, it may be worth considering whether voluntary application could be applied to commissioned scientific outputs i.e. included in the contract. Would the principles of the code be sufficient to address the concerns above? That is not certain, but it is worth noting that the ONS infection survey is run in partnership with unviersities; test and trace statistics are also subject to the regulation of OSR and show very clearly the tension; other management information used in government briefings has also been the subject of OSR interventions to ensure underlying data is published.

    read more

Leave a Reply

Your email address will not be published. Required fields are marked *

*By commenting on this blog you agree to abide by our Terms and Conditions.

subscribe to Fiona's blog