One Year On
As we approach the anniversary of the first national lockdown, I don’t expect any of us would have imagined that a year later we would find ourselves still in it. The events over the last 12 months have had profound and often irreversible effects on the lives of many, and impacted the way we work and how we feel. With regards to scientific development, we have witnessed truly unprecedented times for the agility and speed of treatment evaluations and vaccine development. The daily media spotlight has been on ‘the science’ with a huge public appetite for knowledge and an increasing awareness on the importance of public health research.
In this rather remarkable, disjoint and humbling year we look back on the impact Covid-19 has had on our clinical trials and statistical research. We reflect on what we managed to achieve and what we didn’t. Most importantly we consider the good things that arose as well as the aspects we are looking forward to leaving behind.
Unprecedented treatment and vaccine trials
One aspect that stands out was our delight in witnessing the remarkable Covid-19 platform trials such as RECOVERY and REMAP-CAP. To call them landmark trials feels an understatement. These trials broke many preconceptions around what is possible in terms of size (participants and interventions) as well as the speed to develop, setup and deliver. While we hope we don’t experience the unique circumstances that led to these hugely successful trials again, the existence of them now allows us in trials research to know what is possible.
Notably the generous clinical collaborations that enabled these vital research questions to be answered definitively and within such short timescales. We have already witnessed our clinical collaborators being inspired by these trials, propelled by new ambition. During Covid-19 we took the opportunity to contribute to two treatment trials. While these are dwarfed in scale by the platform trials they did allow us to appreciate some of the personal demands there must have been on individuals team members involved in these record breaking trials. We had a taste of the bumpy road involved when speed is critical and, though inspired, remain realistic about what is possible to be sustained when crisis is out of the equation.
Responding to interrupted trials
Early on we realised the pandemic would impact current trials in progress. Some treatment visits would be missed due to National lockdown restrictions or fears of contracting the virus, whilst participants infected and hospitalised with Covid-19 may no longer be able to provide outcome data. This would likely result in a higher rate of missing outcome data, with new and non-standard reasons for missingness. We therefore devised a four-step strategy to facilitate the handling of unplanned disruptions and missing data in trials due to the pandemic. Our approach involves (i) clarifying the treatment effect to estimate with respect to the pandemic, (ii) establishing what data are missing and why this may be, (iii) performing primary analysis under the most appropriate missing data assumptions and (iv) performing sensitivity analysis under alternative assumptions. As disruptions continue we hope this strategy will help support statisticians and investigators to maintain the scientific integrity of their trials.
Impact on planned meetings
Like everyone else in 2020, we had plans! Once lockdown started we initially delayed these, but soon realised we needed longer term solutions as it became apparent that the pandemic was here to stay. In March 2020 we had been due to undertake our first expert elicitation meeting to inform the analysis of a Bayesian non-inferiority trial, and in April we were due to host a workshop in collaboration with the UKCRC CTU statisticians’ operation group seeking a consensus to support researchers in their choice of visualisations to present Adverse Event data from RCTs. We turned both meetings to remote delivery and, while initially concerned with the loss of interaction in these meetings we found advantages for the remote setting.
For the Bayesian elicitation meeting, potential psychological biases and any conversation monologues were easier to manage. Similarly with the Adverse Event visual consensus we found the chat function allowed participants to raise items during discussions which helped spread conversation and share ideas across members. That’s not to say the remote meetings were not without challenge. There was more potential for distraction and aspects such as intermittent internet connections and simultaneously managing working with caring responsibilities meant that it was difficult for some delegates to take part fully.
The visualisation consensus meeting, originally meant to take one day, was split over three days a week apart to ensure attendees were not overly fatigued and remained energised throughout- the experience from the elicitation meeting endorsed this as a wise strategy! From both meetings we also learnt of the benefits of using anonymous online voting systems to gather opinions and make decisions, helping to ensure attendees had an equal voice. The online format also allowed a greater attendance from a wider geographical area than we had hoped to attract to York and London.
Our cancelled conference talks
Both as individuals and as a team we had conferences and talks we were due to give. We had been looking forward to travelling to Barcelona, Baltimore, Nice, Sheffield, Harrogate, York, Dublin, and Portsmouth in 2020. It wasn’t long before we realised that these were sadly going to be cancelled or postponed. Conferences and scientific meetings provide invaluable ways to promote and share findings of new research, they also provide connections between teams and allow space and time to stimulate new project ideas.
From the initial cloud of disappointment emerged a silver lining as more communities converted conferences online. There was opportunity to attend not only the conferences we had planned but ones we had not. With digital meetings the fees were much lower, and being online allowed us to fit them around our schedules.
The talks we had originally planned were adjusted to virtual presentations including chats, online polls and break-out rooms as a way to provide connection with the delegates. We were also able to communicate emerging Covid-19 statistical research work. Along with the rest of the world we were soon able to navigate features of Zoom and Teams (whatever happened to Skype?).
It wasn’t just our work that was affected by Covid-19, the peripherals of work had vanished. We were quick to set up regular social Teams calls alongside the plethora of work-related Teams meetings to replace our in-person interactions. These ranged from morning coffee breaks to catch up with the team to playing games or doing a quiz. A buddy system was also set up to get people talking to each other during the week, providing an opportunity to take a break within the working day or just talk to someone.
As the months went on, meeting fatigue set in. We saw our initial energetic initiatives tail off. We didn’t fight the more introverted way of working, instead we replaced frequent rapid catch-ups with longer and less structured monthly socials. This allowed us to place emphasis on celebratory events (which usually would have incited a trip to the pub). We celebrated successful fellowship applications (Suzie & Ellie), new starters (Ellie, Leila & Graham), along with farewell send-offs (Lien & Anca). Last summer we were fortunate to align a day we were back in the office, so an unexpected (and oddly novel) socially distanced but in-person celebration was able to occur.
With the long dark nights looming in November we choose to set up a challenge to encourage us to get outside during the day and walk or run around our local areas. The set target was to collectively walk/run the length of the UK. We had a good response to the call for walkers and runners across the unit and were able to smash out original target. The success has inspired us to repeat this in April and this time we are doing Rocket Miles and headed to the virtual Moon!
We have all had to adapt. In the unit there are some adaptions we would like to hang onto, whilst some we cannot wait to leave behind. We are looking forward to saying goodbye to home-schooling, isolated remote working, social distancing and stilted remote conversations (though no doubt on some occasions limiting to only one person speaking at a time can have be an advantage!) We have seen our personal IT skills improve, new online etiquette develop and had much fun from the embarrassing Zoom moments including parents, children, partners and pets, who unwittingly joined and often unknowingly contributed to our meetings. The pets have become such a feature…. who knows, maybe an office dog will be a new addition to our work family when we finally return.
Victoria, Rachel, Suzie, Giles, Jack on behalf of @stats_CI