Yet this approach in medicine is problematic, due to its innate complexity and the consequences that this has on a concept such as objective truth. This problem has been highlighted in the research arena before, and Dr Bearley noted the recent open letter from
Professor Trisha Greenhalgh to the BMJ on qualitative research (http://www.bmj.com/content/352/bmj.i563) and the Medical Research Council (MRC) document on complex interventions (https://www.mrc.ac.uk/documents/pdf/complex-interventions-guidance/). This latter document was brought about by the high number of ‘failed’ quantitative trials, a consequence of searching for hard truth in a sea of confounding factors, and so she asked whether we were undervaluing the role that qualitative research could play. There is bad quality qualitative research out there too, but there are different strengths when comparing it to quantitative methods, and this is not always appreciated.
What is the signal? There is potentially a huge amount of data at the moment and extracting appropriate meaning from it can be a real challenge with risks of misidentifying what the real ‘truth’ is. Transparency in the process is therefore important. An excellent example of this noted by Professor Van Sta is the examination of the relationship between bisphosphonates and oesophageal cancer. In this example we can see 2 different studies published in 2 high quality journals, using the same data set, but achieving 2 different results. (JAMA https://jamanetwork.com/journals/jama/fullarticle/186382, & BMJ http://www.bmj.com/content/341/bmj.c4444) How did this come about? Another example, is the use of the Quality Outcome Framework (QOF) in primary care. Diabetes is a serious condition that needs identification early in the community, and so the QOF provides an incentive to help this process in primary care. This means that the ‘incidence’ of diabetes is increasing, but is it increased identification or an actual increase in the number of patients with the condition? Whilst covering many similar themes, there was also an interesting contrast with the previous lecture. With recognition of the challenge of medical complexity, the question is are we able to use computing power and data size to make it practically applicable? Digital data as an overall volume is increasing but the analysis is hard, with clear potential for misinterpretation. As such, he draws us back to the three R's of data science (as with laboratory science): Repeat, Reproduce, Reuse.
However, research and policy-making move at very different paces. The methodical and careful path of research is at odds with the demands on policy makers to have the answers here and now. This can be at both a national level of policy making, where there may be significant political dimensions, and at a more local commissioning level, where local priorities may have more of an impact, but still in a time critical manner. As such, there needs to be an integration between them. This process of integration is complicated (see picture below) and isn’t working that well. There is often a significant lag between research and ‘front-line’ impact (frequently quoted at 17 years) as well as being haphazard in nature. There is an approach to tackling this though: acknowledge the problem, acknowledge the evidence deficit, aim to get unequivocal results, and promoting research. There is still a certain sense of appropriate timeliness, and areas with higher political weight at a particular time will have an easier ride. There can be a bottom up influence as well though, through the royal colleges, clinical networks,and specialist societies. Indeed, this bottom up approach from driven individuals can promote the gradual change that ultimately leads to a notable impact. Public involvement and engagement are also useful in a similar way, perhaps developing the political incentive. This may arise through engagement with organisations representing patients e.g. special interest groups, or local politicians. An example of such a unified approach can be seen with the ‘academic health science system’ Health Innovation Manchester (https://www.healthinnovationmanchester.com/). Systems such as these aim to bring together the many differing groups discussed to speed up the whole process of implementation.
The first research proposal discussed was by Dr Jaimin Patel on the HOOPES project, looking at the use of high flow nasal cannula oxygen to reduce postoperative pulmonary complications following emergency surgery. The background for the project is that the National Emergency Laparotomy Audit (NELA) has confirmed that emergency laparotomy is a significant intervention for patients, and there are specific high risk groups of patients that we encounter within this population, for example more elderly patients. There is unclear evidence that HFNC can improve outcomes, and the recent Cochrane review didn’t show a clear difference. The aim of this study was therefore to look at the higher risk patients (e.g. age, p-possum, ICU patient) who were undergoing an emergency laparotomy and subsequently going to critical care. The aim was to be as pragmatic as possible, with randomisation prior to extubation to either the intervention of HFNC oxygen for 48h (or until CCU discharge) compared with normal oxygen therapy.The outcomes were primarily 7 day pulmonary complications and 30 and 90 day outcomes, and at this time it is being a feasibility study.
The next talk was from Professor Rupert Pearse on the research project named OSIRIS: Optimising shared decision making for high risk surgery. This is currently at the last stage of assessment to get started and is focusing on the high risk surgical patient. The concern is that there may be a group of these patients who are undergoing high risk surgery but for whom this is not the most appropriate intervention. These patients are often complex with multimorbidity, but discussion about the risks and benefits with these patients is challenging.
As such this study aims to look at the contrast between patients’ opinions before surgery and after – did they make the right choice? Overall the goal is to try and minimise the incidence of ‘decision regret’, and develop tools to help with this.
There are 3 main categories where this will be focused:
1. Major joint replacement – this is not life-saving but primarily aimed at symptom relief
2. CABG – this is life-saving at a population level but not necessarily at an individual level.
3. Colorectal cancer surgery – surgery here is life saving for each patient, but can carry a significant risk of worse outcomes for patient.
The study will start by gathering data on these aspects of decision making, and also looking at more QOL outcomes rather than just mortality at 30 days. Subsequently, the goal is to use this data to design an intervention tool to help with shared decision making, an example being an app that will take onboard which domains are particularly important to each patient e.g. mobility, cognition, and provide understandable data on how they might be impacted e.g. a forecast on the number of people that have impaired mobility. The final aim would be then to test this tool in a clinical trial to see if this makes a difference on this ultimate goal of helping patients make the right decision for them.
The third project was by Dr Joyce Yeung who was discussing the REGARD study. This is a Feasibility RCT to compare regional vs general anaesthesia in reducing the impact of delirium in NOF fracture surgery. It has been a long-term goal to try and improve the outcomes in this group of patients and although there have been improvements in care delivery there is still a way to go. Delirium is increasingly being identified as a bad thing for patients, and indeed is particularly prevalent in this cohort. The pathophysiology is not particularly understood, there is no current cure, it has a notable impact on normal recovery, and survival and morbidity (for instance long term cognitive decline) are both notably impacted upon. In theory, regional anaesthesia is better than general, but the evidence remains inconclusive. The Cochrane review on the topic (http://www.cochrane.org/CD000521/ANAESTH_regional-or-general-anaesthesia-hip-fracture-surgery-adults) didn’t give an answer but a big issue is the dearth of quality, quantative data on the topic. As such, they are doing a feasibility study to see whether is it possible to randomise between techniques (with the challenge of feasibility being a frequent criticism of this area of research). The study started last month (with 11 patients recruited so far) and is looking at how many patients can successfully be recruited, the interventions that are needed e.g. analgesia demands, and subsequently outcome measures. There are some exclusion criteria e.g. polytrauma, absolute CI to regional, as well as the possibility for surgical/anaesthetic opt out. Hopefully, the study will provide evidence that this is an area where a large RCT is actual feasible in order to gain some higher quality data to try and get a clearer answer on this important question.
There is more information on this available at: https://www2.warwick.ac.uk/fac/med/research/ctu/trials/regard/
Dr Roisin Baker delivered the final presentation of the morning on PaTHOS: Prevention and Treatment of Hypotension in Obstetric Surgery. The clinical question is around whether vasopressor choice and mode of administration affects the incidence of maternal nausea and vomiting in elective caesarean section under spinal anaesthesia. It is well recognised that hypotension is common in these patients as a consequence of spinal anaesthesia. In terms of the current research there are 16 trials currently looking at vasopressor use, but these are primarily focused on foetal outcomes rather than maternal ones. The hypothesis is that in elective LSCS, controlled syringe pump delivery of vasopressor will provide better BP control and better maternal satisfaction. The ultimate aim would be to look at producing an evidence based guide on this area, though they are still early in this process. The initial pilot projects have shown that only about 50% of trusts have a vasopressor protocol in place, and there is significant variation in practice. The plan is to work towards this ultimate goal through a stepwise approach including a systematic review of the literature, a survey of current practice amongst anaesthetic providers, discussion with patient groups about the important outcomes, and a observational study looking at the outcomes of patients and the relationship between clinical practice.
Excellent advice was also available from Professor Rupert Pearse, who provided his insights into surviving the approval process. A very abridged analysis of his session could be summarised as: the process is complicated, it is frequently changing, you are likely to not know what to do, but the solution is to work constructively the R&D department. This can be surprisingly difficult when the interactions are occurring with people who have never met before and purely through the medium of the written word (long email threads) but spending the extra time to optimise communication (such as popping down to the office for a face to face chat) can pay dividends.
Dr Anton Krige also delivered a session on ‘How the NIHR CTN can help you?’ (which I was unfortunately not able to attend).
Research Proposals 2
The afternoon continued with the second session on research proposals. First up Professor Rupert Pearse returned with Dr Mark Edwards to update us on FLO-ELA (Fluid Optimisation in Emergency Laparotomy). Dr Edwards, who is leading FLO-ELA, described their study which is looking at cardiac output guided fluid therapy in emergency bowel surgery. So far they have gone through the general pre-trial processes, with the main challenges being relating to data retrieval from national organisations. Data collection has been commenced in September this year, with 57 sites confirmed (hoping for around 100 altogether) and 20 patients recruited so far. They have been using the NELA process form and the base template for this, allowing improved streamlined. In general, the process has been going well, primarily with the challenge of resistance from clinicians with pre-determined opinions about efficacy, despite relative clear equipoise in the literature.
More information is available at http://www.floela.org/
Dr Iain McCullagh was next to speak about PODIUM, looking at Postoperative delirium in the United Kingdom. The James Lind Alliance priorities (http://www.jla.nihr.ac.uk/priority-setting-partnerships/anaesthesia-and-perioperative-care/top-10-priorities/) were again mentioned, with the same questions of focus being noted – how can we improve recovery from surgery for elderly patients? - a driver that had been mentioned in the morning too. Delirium is both unpleasant and harmful, and this harm can be prolonged. It is also common, and the elderly population, who are at particular risk, are increasing in size. The recent systematic review by Siddiqui and colleagues (http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD005563.pub3/full), provided little information on the factors that impact on postoperative delirium. The aim of this study was therefore to try and understand this area better. The feasibility arm is currently underway in the Northeast of England (PODIUM 1) which is a 100 patient feasibility study. The next project is a 5000 patient prospective cohort study (PODIUM 2), with a tentative ultimate goal that an intervention study may be possible. The aim is to assess the feasibility of running the study, and of note the role of involving trainees within the research process. Data is collected on pre-operative factors, intra-operative parameters and follow-up data (incidence of delirium, length of stay). This will hopefully provide information on ‘target areas’ that might impact on the incidence of postoperative delirium (common suspects being benzodiazepine use, depth of anaesthesia). This would hopefully lead on to the development of PODIUM 3, that would aim to test focused ‘delirium-friendly’ anaesthesia.
More information is available at: https://incarn.net/projects-events/podium-1-study/
Dr Robert Bolt followed presenting the MAGIC study (Melatonin for Anxiety prior to General anaesthesia In Children). The background here is that pre-op anxiety is pretty common in children, with current premedication being suboptimal in some ways. The aim of this study is to look at whether melatonin has a role as a safe and effective anxiolytic in this role. There is evidence of its benefit in adult day case surgery, but challenging to unpick in this cohort, although Gitto and colleagues looked at melatonin versus midazolam, suggesting similar efficacy (http://onlinelibrary.wiley.com/doi/10.1111/jpc.13007/abstract). The target group is children age 6-14 undergoing dental and ENT procedures under GA who are assessed as having high levels of anxiety. As well as quantative components, assessing the levels of anxiety and factors such as sedation, they have also included a qualitative component to try and look at possible challenges with implementation.
More information can be found at https://www.pomctn.org.uk/MAGIC
Mr Simon Bach was next to talk about MIDAS (modifying pain using drugs around surgery). The trigger for this was the recognised importance of pain control around surgery, but the known problems with achieving successful analgesia. NSAIDs are recognised as useful analgesia drugs, but associated with specific risks and side effects, and as such there are definitely notable barriers to using NSAIDs in general (renal dysfunction, impaired wound healing, cardiac toxicity). Initial pilot studies have suggested pretty wide ranging practice at the moment suggesting that the benefit versus risk balance seems to be hard to unpick, and a survey of practitioners suggested interest in this being an important clinical study to provide more information. This study is therefore to try and randomise the provision ibuprofen with PPI for 5 days compared to placebo. The primary endpoints are both efficacy (though this is probably likely) and risks of harm (AKI).
Last of the session was Dr Caroline Thomas who presented on the DALES study - drug allergy labels in the elective surgical population. This is the proposed next study of the RAFT group (Research and Audit Federation of Trainees - http://www.raftrainees.com/). The aim is to look in more detail at the nature of allergy status documentation and their actual significance. Drug allergy labels are very common but the actual clinical significance can be hugely variable. There is a suggestion that the label of a drug allergy can have a significant impact on the prescribing practice of clinicians, the issue being particularly well studied in the penicillin ‘allergic’ group. These allergy labels can also prove to be particularly tenacious, and the process of ‘delabelling’ can be very challenging. The aim is to do a 3 day snapshot study using a paper free data collection process on all patients undergoing surgery. The goal is to better define the prevalence of patient-reported allergy status. There will also be an anaesthetist survey to try and explore the knowledge and attitudes of allergy status. This survey will also include a opinion gauging exercise on approaches to improving this labelling.
A final follow-up study will follow up on patients who have a penicillin or opioid allergy documentation and see what has happened to these. Have these labels changed following this clinical interaction? The study is completely observational at this time, with no role for intervening in allergy documentation and the aim is to start data collection in spring 2018.
I would also like to pay special thanks to Dr Andy Martin and Dr Gareth Kitchen from NWRAG who were fellow social media activists at the conference, and to the UK POM CTN team for kindly inviting us to attend and help out with the social media.