TTA Contributing Editor Charles Lowe asks whether it is now time to stop looking back to the UK’s pioneering Whole Systems Demonstrator (WSD) programme.
As the person who led the bid for Whole System Demonstrator status for LB Newham back in 2006/7, this is my case that it’s time now to bid farewell to the programme, as soon as is possible.
Why? This was a great programme that came up with some encouraging results for telehealth, and taught us a huge amount about how best to implement telehealth and telecare. However the echoes from that long gone time are increasingly providing ammunition for the naysayers, when in reality the world is now a totally different place. The technology is unrecognisable from that that we considered when bidding for the WSD in 2006; it is far more efficacious and far cheaper; and it can be deployed much faster & for many more conditions, opening up many possibilities not available to us when we won in 2007. We now know much more about how to implement the technology too: in particular it delivers greatest benefit when a part of an overall programme for improving care and not, as the WSD randomised control trial (RCT) treated it, as a simple intervention, like most drugs. In retrospect therefore there were significant weaknesses in the way the trial was run.
The continuing drip-feed of WSD results is sadly resulting in a diametrically opposite result to that intended from the WSD: it is slowing the development of telemonitoring in the UK, and elsewhere. Perhaps the most recent example of this is the publication by the Royal College of GPs of their 2022 GP: a vision for General Practice in the future NHS, where the Compendium of Evidence dismisses telehealth (P36) on the basis of a WSD paper, leading it on the following page to quote remote delivery of test results as the most advanced medical development likely to affect GPs over the next nine years. At every conference too – including the recent excellent Kings Fund International Congress on Telehealth & Telecare – there are clinicians who after listening to a WSD paper appear to believe it to be the current state of the technology and begin sounding off about wasting money.
So what are the reasons? Let me begin with equipment specification and cost. In the autumn of 2006 we made what I still believe was, then, a great choice of telehealth equipment that we would use, were we to be successful with our WSD bid. In the intervening seven years however, equipment cost has fallen substantially and functionality has improved significantly.
A related issue is that the WSD was in a sense an oxymoron: as both control and intervention groups were in the same (Newham) system, we were unable to demonstrate the effect of whole system change. Worse was that we had to run two systems side by side – one for those in the trial, and one for the rest, which naturally increased operational costs. The result of this, combined with the WSD equipment costs, means that the historic costs, which were far higher than a well run programme now results in, are of value only to historians.
Looking next at flexibility and speed of response, perhaps most importantly the prospect of telehealth as an app downloadable onto a smartphone both significantly further reduces upfront cost and completes the move to a mobile, instantly available, service – compared with the month that our telecoms provider used to require to install a broadband line (and then only if the application form was completed perfectly first time). So, for example, equipping someone with telehealth to facilitate early discharge from hospital is now easy, whereas in 2007, when the trial was being designed and the types of interventions finalised, it was not even seriously contemplated.
Likewise, instantly installable telecare was often not available then either, to provide monitoring cover as soon as a user was assessed for it.
Those who have not endured a rigorous RCT will perhaps be unaware of how restrictive communications with patients can be, both in the patient selection phase, and in interactions once users have been allocated to the intervention or control groups. It was therefore especially galling when I chaired a session at the recent Kings Fund event to hear a paper showing no significant evidence of self-care behaviour in the intervention group when that was not one of the principal WSD objectives, and the RCT’s rules of conduct expressly forbad the sort of interaction that would have fully encouraged patients developing self-care behaviour.
Patient selection is also worth a mention. In Newham we unintentionally ended up removing many of the neediest people from possible involvement in the trial. This happened because, at the time of the award, we had already rolled out almost 3,000 telecare installations. Although this total, which represented a little over 1% of the Borough’s population, obviously couldn’t include all those with the highest need, it certainly contained many. When a decision was taken after we had won that a randomised control trial (RCT) methodology was to be used, of course all these higher need people were then excluded from the WSD trial because they already had technology.
It’s fair to say that the recruitment process for the initial 3,000 had added further bias too, because those already paying £2/week for our pendant-only community alarm who spotted that, by moving to full telecare, they got a better service, and one that at that time was free, did so in large numbers and so excluded themselves from the trial. Only those who continued to pay £2/week for the restrictive service were therefore eligible for the WSD trial. Also excluded from the trial were the two categories of telecare user who we considered at the time to deliver the greatest benefit to the Council – people with a learning disability, particularly those who required ‘waking night’ care, and people caring for those with dementia that were prone to wander.
Perhaps as a consequence of the above, the headline results were disappointing to those of us who have seen hugely impressive results from using technology as part of a genuinely whole system intervention, although of course using a much less rigorous evaluation method (typically before-and-after). Combined with the very high historic costs, that gives the WSD extremely unfavourable cost/QALY figures.
So what to do? I am told that now that WSD funding has ceased, academics are producing WSD papers in their own time so there will likely be a continuing stream of such unwelcome results for a little while yet. Of course it’s understandable that academics with access to a huge quantity of historic data may have different objectives to those of us keen to implement modern technology successfully. Perhaps the DH could though be encouraged to provide just a little extra funding for them to finish off their current WSD work quickly? This would draw a line under the whole affair so the world can move on, looking for ways of using technology best, rather than looking back to prove yet again that we were still at a much earlier stage in technology use six (shortly to be seven) years ago?
None of the above is a criticism of the WSD trial – it was a visionary approach to kick-starting the use of telecare and particularly telehealth to improve patient outcomes and reduce costs. Partly as a result of the trial, equipment cost and functionality have changed so much, and we have learned so much more about how best to deploy and prove the benefits of the technology. Indeed, in spite of published material, all three WSD sites decided subsequently to mainstream telehealth which is as good an indication as any of its value (all had already mainstreamed telecare) – in Newham the business case was primarily founded on improved efficiencies in the use of clinical staff rather than any of the results of the WSD Trial.
Good points, well made Charles. Even in the pharmaceutical world there is tacit acceptance by some that the RCT model isn’t always most suitable, especially in the fields of personalised medicine, pharmacogenomics etc. And, as you point out in your other article, designing clinical trials for medical devices is even more problematic. Like you, I believe that there is a growing body of evidence from other sources which should now take the limelight from WSD. For example, the Portsdown Case Study by Medvivo, featured on p.9 of the Summer edition of TSA’s Link magazine, demonstrates topline results from the first 6 months of the following reductions: 85% in GP home visits, 57% in unplanned admissions, 67% in GP appointments, 52% in A&E attendance. I agree with you that, since telehealth technology and service provision has moved on since WSD, the focus of debate should too.
In Wessex although not one of the WSD pilot sites, Southampton was one of the 12 action learning sites linked to the project. In the paper describing the outcomes from these learning sites Giordano, Clark and Goodwin highlight three issues as being key when implementing telehealth and telecare: leadership; working practices, skills and development; and data management. In the work the Wessex Health Innovation and Education Cluster(HIEC) has done in the past two years helping to support NHS organisations and local authorities to implement telehealth and telecare the above issues have been key. Where there has been strong leadership, sufficient project management capacity and a commitment to telehealth/telecare as a tool in the way care is going to be delivered progress has been made. Staff need to understand what is expected of them, training courses and awareness sessions have helped to raise the profile of technology as an enabler. The Wessex HIEC has created a series of training tools that are available on the Wires website and these are available to staff across the UK.
Finally the issue of data management cannot be underestimated, the trusts that are making the most progress are those that have tackled this head on and have joined up the various data sources that they use into a comprehensive system that can support clinical decision making. Some trusts have opted for a shared risk and reward relationship with their IT supplier and this seems a good option when ways of working are changing so rapidly.
Drawing a line under the WSD results would help the HIEC to continue to make progress as these results are still raised at meetings particularly with GPs. They can turn the mood of the meeting away from considering how telehealth and telecare are going to enable organisations to continue to deliver quality care to growing numbers of people living with long term conditions.
Language matters. And some people have thrown the e-health, e-mental health and m-health baby out with the telecare bathwater. Now where is spotty dog to save the day?
I worked for Charles in Newham, both to set up telecare and on the WSD Programme. I subsequently took over responsibility for the mainstreaming of telehealth services first for Newham PCT and then, as WSD Legacy & Transformation Manager, for East London NHS Foundation Trust.
I can only support everything that Charles says in his artcile, with one exception. The business case for mainstreaming telehealth by our provider organisation did not ignore all the WSD results, rather it took one that everyone thought of as negative and showed that in fact it was positive. The fact that there was no improvement in quality of life also meant that there was no detrimental impact through the use of telehealth to provide care and as our clinical staff could support four times as many people with telehealth than they could without there was a clear efficiency case to be made.
One issue with the NHS is that its default method of working has been on paper and face to face. The need to move away from the first of these methods is now widely accepted but more work needs to be done to convince service providers that supporting patient self care and self servicing using channels other than face to face is practical. Telehealth is one part of the solution. The fasting growing telehealth solution in Newham is the NHS’s “Florence” system that uses the patient’s own mobile phone and SMS texting. Equipment cost – nil, patient training cost nil, setup time 2 mins. Our Diabetes Specialist Nurse Team use both Skype and Florence to support patients, obviously not appropriate for everyone but this is no reason to deny it from those who really appreciate being supported in this way. Flo was not available for the WSD Trial else we would have used it and the results in terms of cost would have been very different.
Charles, I believe it is important to put the WSD in its proper context, but I don’t believe we should bid it and the knowledge gained farewell. Despite the fact it did not produce the results we were hoping for, it did provide many learnings which we can use.
I agree the process of drip feeding the information is a problem. We need greater transparency around the results and a more robust discussion around next steps.
One huge thing still missing is clinical engagement and involvement. Without it, as we saw in many of the WSD projects, you cannot have a successful healthcare transformation project.