The Sunday before last, Fi tweeted about the possibility of starting a journal club on Twitter to discuss a variety of medical papers. Within a week interest grew, a paper had been chosen, Fi and I had set up a blog, a Twitter account and last Sunday saw the first “meeting” of the Twitter Journal Club. The response was overwhelming – in my wildest dreams I could not have hoped for such a successful first discussion. The paper we had chosen was by no means a simple choice – Rivers et al is a complex paper on critical care but the level of critique more than met the standards of this challenging paper.
The Critical Appraisal Skills Programme (CASP) guidelines were used to give a framework to the discussion (developed and produced by CASP at the Public Health Resource Unit, Oxford). This toolkit gave a clear structured approach to reviewing this randomised control trial and was a fantastic tool.
It is an impossible task to try and summarise the nearly 2 hour discussion into a single blog post, so instead I have decided to pick the five take-home discussion points:
1. Randomised control trials and written informed consent
The RCT bit worries me…just how do you get “written informed consent” from someone in SIRS…? Truly no influence…?
When critiquing a paper, and especially a trial, it is easy to overlook the ethical issues in favour of concentrating on methodology and whether the statistics and results stand up to scrutiny. This paper recruited patients with severe sepsis and septic shock and the issue was raised – can we ever really get written informed consent from such acutely unwell patients for involvement in a trial? This trial had approval by an ethics committee although some felt that the issue of consent had not been explored fully enough.
2. Methodology – the issue of blinding
This trial was interesting in that it was partly blinded: clinicians initially treating the patients were unblended; however, once care was transferred to the critical care team, the clinicians who took over care were blinded. This lead to an interesting debate:
Full on double blinded RCT ? no. I can’t think of how they could do better to answer the question & care for patients ethically
The consensus can be summed up by the tweet below:
not having this study as double-blinded doesn’t reduce the quality of its findings
3. Headline result
The reduction in in-hospital mortality was reported in the paper as the primary outcome – this was used to calculate the number needed to treat by several participants:
If 16% ARR in in-hospital mortality, that’s a NNT of almost exactly 6. (No calculator to hand)
This figure had an impact on several of the participants and highlighted just how effective the interventions in the paper had been shown to be.
NNT of 6 is very low!
4. Single Centre Study: Does this reduce the impact of the paper?
The reduction in mortality in the intervention group was a staggering figure; however, many raised the issue that this was a single centre study – did this limit its impact? The issue of resources was also raised: different hospitals in different parts of the world allocate resources in different ways – does this make the Rivers et al paper less relevant?
Just saying that only studying single centre limits default wider application
@TWSY @twitjournalclub I think resources are a key point – research is useless if can’t be applied.
One point raised, later on in the evening, was the differing nature of A&E / the ER in the USA compared to the UK, with questions raised over how this might affect the relevance of the paper to UK clinicians. Furthermore, while its impact has been huge, is this paper relevant in terms of resource-poor areas?
5. Rivers et al and Surviving Sepsis
This leads onto my final take-home point from the discussion – the impact of this paper. Many comment on how important this paper has been in terms of impact and the way we now manage patients with sepsis, particularly with the Surviving Sepsis Campaign.
The Surviving Sepsis Campaign – the impact of care bundles is clearly seen in everyday practice
Time is of the essence when managing septic patients, the importance of the “early” part of “early-goal directed therapy” should never be underestimated:
In real life we’re all (or should be) identifying & treating septic patients earlier than used to, and getting them to HDU
As far as we can tell this is the first journal club to be set up on Twitter and run exclusively through Twitter. While others have “live-tweeted” from real-life journal clubs in the past, one key benefit of hosting a journal club on Twitter is that it allows all sorts of people to participate, from students to consultants. Particularly for students, this can provide a fantastic learning opportunity.
A criticism of our journal club has been that the concept would be difficult to follow. However my experience is of the opposite, there was a huge amount of interaction between all the participants and this lead to a lively and interesting debate.
A static transcript will never be able to show a dynamic process well but as the weeks go on Fi and I aim to improve on this first meeting and on how we present the discussion as a summary on the blog. This is a huge learning curve for both of us and we appreciate any suggestions on how we can improve.
Thank you to all who participated and I cannot wait for the next discussion on Sunday 12th June on Geoffrey Rose’s ‘Prevention Paradox’ officially titled ”Strategy of prevention: lessons from cardiovascular disease”, available here.
Edit: Just a reminder that the transcript from Sunday evening is available here, should you wish to look through it. Also, please do comment on this post and continue the discussion about the paper if you have any other points you’d like to make.