An interview with Professor Martin Gibson, NorthWest EHealth Chief Executive, conducted by April Cashin-Garbutt, MA (Cantab)
What is Linked Database System (LDS) technology and why has it been developed?
LDS stands for Linked Database System, which essentially describes what it does. Our LDS technology uses multiple sources of information from hospitals, GPs and pharmacies. However in the future this information could also come from other health data sources e.g. your smartphone, Fitbit etc.
The LDS is a platform that allows information to be gathered from all these different sources and puts it all into one place. It can be used to analyze the outcomes of clinical trials.
We started developing the LDS for the GSK (GlaxoSmithKline) sponsored Salford Lung Study (SLS). Development began in 2011 and over the years we have continually updated and improved the technology.
NorthWest EHealth (NWEH) is based at the Citylabs, which is a biomedical centre of excellence, located in the heart of Manchester's Innovation District and within the largest clinical academic campus in Europe.
How does LDS technology integrate with patients’ electronic health records and what capabilities does this enable?
Everyone’s health records are spread across multiple systems and places. For example, medical information will be held by GPs, hospitals and pharmacies, and currently these records aren't connected together. LDS takes the information from all of these very different kinds of systems and compiles it into one record, which can be used to monitor what's happening.
However, it's a bit more sophisticated than that. As we are using it to run clinical trials we have to be absolutely sure that when we take information from one of these different places and move it around and process it, that we haven't changed the information in any way. Therefore, we validate every single part of the process. That makes it considerably more complicated but it means that you can absolutely trust the information you’ve got.
How do you make sure the information is secure?
We keep data as secure as we possibly can by using encryption and keeping everything behind the NHS firewall. All the data within the LDS remains within the secure NHS N3 network and is afforded the same security as patient healthcare records. Access to this data is tightly controlled. We also carry out penetration testing, to check for vulnerabilities in our systems that could be exploited. Having ISO 27001 certification at NWEH also means that we have robust security standards when handling patient information.
Can you please outline how LDS was used in the ground-breaking Salford Lung Study?
When we wanted to set up the Salford Lung Study the whole idea was that we’d have as little interference as possible into the everyday lives of those participating – and that’s really different to a normal clinical trial, which can be quite intrusive with multiple visits to clinics.
We wanted to be able to see a person at the beginning of the trial, get their consent and all the things that you would normally do, and then not have to see them again until the end of the trial, which was 12 months later.
Over that 12-month period the person could be receiving a medicine that was unlicensed at that time, i.e. it was still in the clinical trial phase. So we had to be absolutely certain that we could monitor the safety and the outcomes in close to real time. That's why we had to build LDS as there was nothing in existence that could already do that.
To be perfectly safe we needed information from all of those different sources - GP, pharmacy, hospital - to make sure that every time somebody touched the NHS, we knew they were okay or we knew if they were having problems, and we could do something about it if they were.
What we built was in operation from day one, but we've refined it considerably as time has gone by.
Why was the inclusion of a broad population of patients so important for this trial?
You hear a lot of people talking about ‘real world’ studies. Our LDS technology links consenting patients’ electronic health records (EHRs) across all of their everyday interactions with pharmacists, doctors and hospitals, enabling the collection of real-world data, but with minimal intrusion.
To give you an idea, the first trial that we published in the New England Journal of Medicine in September was on Chronic Obstructive Pulmonary Disease (COPD). COPD is a condition that affects a lot of people, but if you look at standard kinds of clinical trials for COPD and at the inclusion criteria and the exclusion criteria, then only about seven per cent of people that have the condition would ever get into a typical COPD clinical trial.
That means that more than ninety per cent of people who will eventually be prescribed the drug have never been exposed to that kind of medicine – so standard clinical trials do not reflect real-world patients.
We simply didn't know what benefits most people with COPD would get from this medicine. The idea of this study was to show that the medicine was effective across all the different kinds of people that have COPD. We wanted to show how effective it was in a whole population, not just the seven per cent that might get it in a standard clinical trial setting.
It has been reported that LDS technology could enable drugs to get to market faster. What is the reasoning behind this claim?
This is definitely where the technology is heading. What we've discovered by getting information directly from the medical records is that there's no transcribing of that information. It’s not necessary for somebody to sit down and take information from the GP and copy it into a different report form and then submit it. There's a lot of time saved doing that.
The other thing we've learned is that the safety monitoring system that we built for this is actually much better than anything we use right now. It’s much faster so we can receive messages in almost real time, telling us if people are okay. This gives us confidence that we could use this type of trial methodology much earlier in the drug development program allowing us to blend and adapt the current typical phase one, two, three, and four programs so that we could shrink the whole timeline.
It has also allowed us to design trials in a different way. A lot of people talk about adaptive trial design – whereby the design can be changed over the course of the trial. You find the patients who are getting the side effects and then refine the study so that you focus on the people that are getting the greatest benefit and the least problems with the medicine. Then at the end of the studies you have a population of people who would really get the greatest benefit from the medicine. Our technology is very supportive of this kind of design.
In what ways does LDS technology enable rapid patient recruitment?
Well, it doesn't, is the simple answer to that. We have another piece of technology, which we use alongside LDS that is called FARSITE. FARSITE is actually an acronym; it stands for Feasibility And Recruitment Systems to Improve Trial Efficiency.
You can think of it as the front end of LDS. It's like an app that you bolt on that allows us to look at anonymised medical records and put things in categories. For example, if I wanted to find people between the ages of 40 and 50 that have a certain range of blood pressure and are on a particular group of medicines then FARSITE will tell me precisely how many people meet those criteria.
FARSITE helps design your trial, and then the clever bit is that it can then send a link back to the GPs. The GPs can press some buttons and it re-identifies the patients, but only to their GP. Then, because your GP provides all of your direct care the GP can decide, with you, whether or not you would like to participate in the trial. If you are interested, further information and full consent to take part will be obtained in the usual way before anything else happens.
FARSITE provides us with a very rapid way of designing a protocol, finding patients and recruiting people via their usual family doctor.
How could LDS reduce trial costs?
As I mentioned earlier, it saves a lot of time. Perhaps more importantly LDS provides a very different way of doing things. For the two Salford Lung studies together, we recruited about 7,200 patients.
Now, most studies of that size would be done across the globe requiring five hundred or more sites to recruit that number of patients. You can imagine the logistics behind getting all of those sites up and running, making sure all the medicines are in the right place all around the world.
With LDS everything is done in one community of care, in one geography. I don't think anybody has done that before and certainly not of this size. That immediately gives you cost savings because everything is in one place. The logistics are much better and the speed of recruitment means that you can focus everything on just that space. Again, it shortens the timelines and it makes it cheaper to do.
What other technologies are currently available to support the collection of real world evidence and what do you think the future holds for research in real-world settings?
Well, you'd expect me to say this, but this is the future. I'm convinced by what we have done already that this is a better and more efficient way of conducting clinical studies.
Say you wanted to study a medicine that relieves pain and you wanted to know if people were able to get out and about more and how often - you could get permission to link to their mobile phone GPS and you could see whether they're moving around. You could link in their Fitbit (or similar device) and you could see whether they're able to do more. You could link it to multiple household items to see when they're opening or closing the fridge. There's a whole load of stuff that is now becoming available that would allow us to get far more detail on any potential benefit, or otherwise, of a medicine in the real world.
This approach provides us with much greater detail of how things are happening in somebody's everyday life. We've never been able to get that before in a clinical trial. It would really tell us a lot more about the true value of any medicine and that's what people want to know. Often we do trials with particular endpoints that have been pre-determined. In my case, because I look after people with diabetes, it’s likely to be how much does the drug lower somebody's glucose level?
Whilst that may be very interesting from a scientific point of view, from a patient’s point of view it's not that exciting. What they really want to know is – is that going to make me feel better? Am I going to live longer? Am I not going to get all the problems associated with diabetes? Will it mean I can drive my car, go to work, live life like people that don’t have diabetes and so on?
With this new approach to doing trials we'll be able to look for the endpoints that really matter to patients and to the health service.
In what ways do you think technology could help to reduce clinical development time and costs further?
Let's say we could reproduce what we've done in Manchester and Salford at a series of nodes around the planet. These may be places of similar sizes or cities with a population of two or three million around the globe. Applying these would mean you could run the majority of trials in fewer representative settings.
By focusing resources on those areas we could maximise the amount of information we were getting. It would be less expensive as well as much quicker. Because of this medicines would get to market quicker and patients would get them sooner because it would take less time and less money.
Hopefully any savings would be passed on to the payers - in our case, the NHS. It would also mean that the company that produced it could be more assured of the safety of their products and receive an appropriate price for it. As it would have been done in a shorter time, this would mean their effective patent life would be longer. All of this creates a win-win for patients, payers and producers. This is why I think this is so exciting to everybody.
This is not just about medicines. You can use the same technology for evaluating medical devices and for new kinds of diagnostics. It would also work just the same testing new clinical behaviors and how best to embed research findings into everyday care. From a clinician’s perspective it is something I’m very excited about because historically we’ve been really slow to adopt new technologies in the NHS.
What we have developed and continue to create here at NWEH covers the whole spectrum - from the first idea about a concept of new medicine or a device, all the way through to it being used in everyday care and making sure that it's safe and effective. It’s very different to what we do now and potentially very disruptive – but in a good way.
Where can readers find more information?
About Professor Martin Gibson
Martin Gibson is Chief Executive of Northwest EHealth, Director of the NIHR Clinical Research for Greater Manchester and Research and Informatics Director for the Greater Manchester AHSN. He is a consultant physician specialising in diabetes and lipid disorders at Salford Royal NHS Foundation Trust where he was formerly R&D Director of both the acute and primary care Trusts. Martin is an active clinical trialist and has had a long-term interest in the use of electronic clinical data systems to improve healthcare and facilitate research.
NorthWest EHealth (NWEH) is an organisation formed from a partnership between the University of Manchester, Salford Royal Foundation Trust and Salford Clinical Commissioning Group, that specialises in developing supported easy-to-use software tools for clinicians, scientists and managers to undertake research, audit, commissioning and care – http://nweh.co.uk/
The National Institute for Healthcare Research (NIHR) networks provide infrastructure support for academic and life science industry clinical research across the country. These actively managed networks have significantly increased the capacity, capability and performance of the UK across all phases of clinical research – www.nihr.ac.uk.
The Greater Manchester Academic Health Science Network is one of fifteen AHSNs that provide national coverage. AHSNs have a remit to improve patient and population health outcomes by translating research into practice and to develop and implement integrated health care services – www.gmahsn.org.
Renaissance of "food as medicine" in modern clinical trials