The future of patient portals – the decisions in design and tech implementation we take now, will critically impact the future of healthcare delivery
For our latest HTN Now panel discussion, we were joined by Manzoor Ahmed, patient and public involvement leader at Royal Berkshire NHS Foundation Trust; Adrian Byrne, ex-CIO of University Hospital Southampton Foundation Trust and ex-chair of the national CIO Network; and Dr Gege Gatt, CEO at EBO.
Panellists discussed a range of topics around the future of patient portals, including future directions, technologies, challenges, and more.
Manzoor shared a brief overview of his background and experience in this area, including his work as a patient and public involvement lead, “one of the things I’ve been heavily involved in since 2002 is the introduction of a patient portal, and my insight comes from working closely with patients, so I hope I can bring to this discussion some of the things that patients want to see in this space.”
Gege talked about his work with EBO across eleven countries, and how this offered him the chance to “compare and contrast different national strategies in terms of engagement”, as well as his work as a WHO reviewer on digital informatics strategy, creating the digital health information platform strategy for Europe.
“EBO is serving sixteen NHS trusts at present, with a particular focus on patient engagement, and our philosophy is solving the pain points of access and demand.”
Adrian highlighted his experience as a former CIO at University Hospital Southampton, and as national chair for the NHS CIO network, and how this had granted him plenty of insight into the patient portals space.
Learnings and insights on patient portals
Our panel began by considering some of the insights and learnings from previous projects and experience on patient portals.
Gege talked about his observations around the “lack of a personalised touch” within current patient portals, suggesting that in the future, this could be something that technologies like AI could help resolve. He added that in his experience, many patient portals “essentially handle things like tedious forms and pushing out basic SMS”, but “they lack empathy and often engagement is very poor”.
“Most academic data, both in the UK and also in Europe; shows that around fifty percent of patients that visit a traditional portal bounce off, with almost no interaction. Our learnings have been that we need a new approach, or a second generation portal that really brings the conversational dialogue between the patient and the trust to the forefront.”
In terms of work with NHS trusts, Gege shared with us some details on implementations at Somerset NHS Foundation Trust, including an “interesting layer” of the solution which was based on patient-led appointment management and automation with the smart rebooking of appointments. From this work, he continued, his team are seeing “patient satisfaction rates as high as ninety five percent”.
“I believe that comes from the fact that we’re inviting patients to get involved in that dialogue and that communication with trusts, and that’s very different from filling in a form. We’re saving around six hundred hours of staff time per week, because there is a very laborious appointment management dependency in most clinical settings, which I believe needs to be automated. It’s also interesting to see in this project that forty-two percent of all interactions took place outside of office hours, which perhaps reflects that patients want to be able to engage at a time that’s convenient for them.”
In a separate project with Betsi Cadwaladr University Health Board, Gege said that his team had put in place a layer of their solution which “was communicating constantly” with patients on waiting lists, “assisting patients to wait well”, but also “constantly validating that list of patients as a continuous strategic exercise”.
“We found that for every twenty thousand patients that we were validating and communicating with through AI, we were saving around £137,000 when compared to the previous mechanism of doing so. We scored a patient satisfaction rate of ninety-three percent, and around ninety-seven percent of all conversations and dialogues that patients started were fully completed all the way down the expected clinical and patient pathway.”
“That’s really the key”, continued Gege, “is using technology to help healthcare providers to build these meaningful relationships with their patients at a high level of adoption. This is the fusion of technology; of artificial intelligence and traditional patient portals; it’s about creating this new digital front door”.
Manzoor discussed his insight from Royal Berkshire, referring to the debate around whether solutions should be focused on the technology or on being as patient-centric as possible, adding, “in my opinion these approaches are different”.
“Patient-centric solutions are really built around what the patient needs, and it’s about realising that there is more than one type of patient, and you may have different needs for expectant mothers, older patients, and so on. On the other hand, you sometimes have companies developing a technology that will be introduced in primary care, in secondary care, or in tertiary care. That can be difficult to adapt to the environment where it’s being implemented, and we saw those challenges for ourselves.”
Another challenge, according to Manzoor, is around the user interface, where healthcare has “a long way to catch up” with what else is available in other industries.
“The user interfaces are often clunky, and the information you need isn’t readily or easily available, and there isn’t always someone to talk to if you get stuck. Modern patient portals need to be more user friendly, and I think this is where we can increase adoption. Also, introducing new functionality and features is important, so it doesn’t become stagnant, and involving patients right at the beginning is integral.”
As we move into “an era of virtual wards” and Hospital at Home, “the patient portal needs to come into the modern age”, being capable of showing live data, he added.
The future of patient portals
We asked our panellists what they thought patient portals would or could look like in five years’ time.
Adrian considered that “there will clearly be a lot more intelligence in this space in future”, adding that “to date, we’ve been very basic”.
“We’ve been thinking about portals almost just as a way of presenting information to patients, and really there’s no great benefit in that for anybody. If all you could do in your banking app was view your balance, you wouldn’t be that interested. We’ve moved on a little bit, so we’ve got some basic messaging functions, but in the future I hope we’ll see a lot more intelligence being brought into that.”
One of the main barriers to moving forward in this way, Adrian continued, is in integrating the information back with the patient record, and in how, for example, a hospital medical record can interface successfully with a patient portal.
“That is a really big hill to climb for a lot of people, and I think that will continue to be the case, honestly, because lots of systems are very closed in terms of their architecture. We almost need another generation of healthcare systems that are built with these things in mind. But in future, I think it’s clear that patients will be able to manage their conditions a lot more in terms of self-help, and they will be a lot more personalised.”
Gege said that in his opinion, “the era of static, one-way patient portals is over”, and we’re “on the brink of a healthcare revolution where artificial intelligence can really breathe life into every patient interaction”.
“Imagine a clever combination of a super-intelligent chatbot and traditional portal functionality. It meets patients with understanding and empathy, and can recognise any language being spoken, recognise your needs, and guide you through the healthcare journey. I think this conversation isn’t about technology, it’s about making healthcare human again, because traditional patient portals have dehumanised the service.”
The future, according to Gege, has two aspects: firstly, patient portals need to be communication-centric; and secondly, “we need to call things by the impact and outcomes they have, like a patient experience solution, because it underscores the need we have to prioritise the end goals and the human experience”.
On what the two-way communication part of this might look like, Gege elaborated that “interaction should take place via any channel or device a patient chooses, like a device in a car, through an app”. He also talked about the possibility of integrating a virtual assistant into the NHS app which would be “fully integrated with the NHS login solution”.
“The other thing is that a dialogue is a bi-directional exchange within guided parameters. This means that if a patient is filling out a form and they need to ask a question, the AI needs to understand their level of literacy and adapt in real time for their needs. Parts of the conversation which go unsaid, such as the context and the emotional state of the patient, should also be fed back in to the patient record system.”
The interface should adapt to the user, not the other way around, Gege clarified, and should achieve a “positive connection”, rather than “merely completing the required task or exercise”.
Recommended by LinkedIn
“We do that in three ways: personalisation, so a one-size-fits-all approach doesn’t work, and so every conversation or experience needs to be personalised; restoring empathy, because we need to become better at understanding the emotional state of the patient; and lowering the bar of digital literacy. Research published last year shows that across Europe, around forty-six percent of citizens don’t have the basic necessary skills to interact with web solutions, so we need new layers of engagement to ensure that more people can receive the care they need, when they need it, and with fewer barriers to entry.”
Manzoor agreed that moving into a world where AI can help to solve problems and overcome barriers is “something we all want”, but added “I think we have to step back a bit and understand that some trusts don’t even have an EPR system, or their infrastructure is not up to standard”.
“In an ideal world, in maybe ten or twenty years’ time, maybe we’ll see AI playing a greater role, but I think to have a greater level of adoption, we need to have simple solutions which are built around what staff need, which will improve the way they work. I’ve spent time in wards where new technology like a new EPR system is introduced, and you can say, ‘well, that was adopted successfully’, but the staff continuously struggle to use that technology. It’s the same with patient portals, we need to think about what capability our patients have.”
“AI is not the complete solution,” Manzoor continued, “because I think as human beings we still need that human touch”.
Many of the comments left by our live audience echoed some of the themes introduced by our panellists, with one commenter saying that, “change management is key to new models and can then be layered with Analytics/AI and predictive models”, and another considered how “in the future there will be much more functionality” available by going through the NHS app.
One commenter felt that “the future could be very exciting”, provided that “we make patient engagement the front and centre of any development”, adding, “I think it would also enhance person-centred care and would significantly empower patients, which is invaluable from a therapeutic alliance point of view”. They added that this would “entail significant training of patients and clinical staff, as the level of unawareness for digital among clinical staff need to be addressed”.
Challenges
A question from our live audience around new roles required for digital workers and removing the burden from frontline staff to support service users in understanding and managing the move to digital systems, was an opportunity for our panel to consider some of the challenges around patient portals, both now and in the future.
Gege highlighted the importance of technology being seen as something that can augment human capability, rather than replacing it, adding: “The future of work is not one in which AI replaces clinical activity, but one in which AI automates repetitive tasks which are burdensome on human activity. The NHS app is fundamental to our future; it is clear that the investments being made there need to be leveraged, not replaced. The more we can automate end-to-end, the more patients can achieve, the more agency we restore to patients, and as studies show, the higher the level of engagement, the better the healthcare outcomes, because citizens get more invested in that particular care plan.”
In terms of skills, Gege stated that “in the healthcare industry, employability is less about what you were trained to do at university, and more about your ability to change, learn, and adapt over time”.
“I think in the next few years, we’re going to see NHS trusts employing more data scientists; I think we’re going to change population health capabilities to focus far more on data science predictions. We’re going to start asking ourselves how these conversations fit into our clinical processes, and how our clinical processes fit into our interoperability strategy for data transfer. These new skills will be hard for the NHS to acquire because the commercial sector is keenly attracting that type of talent, but I think the UK government has really put policy first and is investing in this type of skill development in a number of ways.”
Another challenge Gege mentioned was around health equity in patient portals, which he stated “we don’t have right now”, as “generally, traditional portals are still quite exclusive”.
“We need to introduce technology that is designed to serve everyone, ensuring that nobody is left behind; that is what the democratisation of health information is all about. That requires a step change from the decisions we are taking within NHS trusts, for us to move into the third stage, which is AI-driven data analysis, because that information we glean from these conversations with citizens will empower us to make smarter and faster decisions, and that can be very transformative when it comes to public health policies.”
Adrian talked about how “unless you manage a situation, it develops almost organically, and what we’ve seen to date is the lack of any real strategy or control in this area”.
“Things have developed, really, based on what suppliers wanted to do in the market, so you have a situation right now where patient portals are mostly delivered as a way of providing access to a medical record. Many of the hospital EMRs now have their own system, where patients can log in and do things with their record in that hospital, and ultimately I don’t believe that’s a good strategy, because this means that your record effectively becomes organisation-centric, and is not based around you, it’s based around what’s been going on with you and a particular organisation.”
Records need to be portable, Adrian continued, “so that you can have conversations with different health professionals that you come into contact with; the record itself needs to sit outside of the organisations that you’re dealing with”.
“Fundamentally, we missed the boat on this, because these things have now already been set up. All of these portals are now stuck on the front of an electronic medical record, and it’s going to take us a few years to dig out of that hole.”
Gege agreed with this sentiment, saying that “we need to think of our healthcare solution as having a number of layers; data is separate from clinical workflow; clinical workflow is separate from the patient touchpoint; and the touchpoint needs to be focused on engagement”.
“Each of these layers is interoperable, and each of those layers is built with specialisations in mind. At EBO, our intelligent patient portal also has another layer, which is an automations layer, where we tackle things like elective recovery automation, digital consent for immunisations, alerts, notifications, appointment management, and so on. The way we look at it is that the enabling muscle that artificial intelligence provides is not just an opportunity for reform, I think it’s a responsibility that we have towards our citizens, towards our customers, towards the various stakeholders and markets which we serve.”
We saw a high level of engagement within the live chat around the topic of challenges, including one comment saying: “It feels to me that the biggest issue really is how fragmented the NHS is – the kind of fundamental change that you’re describing needs a coordinated, almost mandated national approach. The marketplace is so busy, and every division/trust/GP practice can make their own choices on which solution to go with. Without a completely different approach to change in the NHS, really implementing impactful change described is very hard to see.”
Another commenter pointed to “the fragmented approach of integration” as “the key blocker in the NHS”.
Tackling challenges in the short-term
As our panel discussion drew to a close, we asked our panellists what challenges they would solve if they could choose one to really make an impact in the short term.
Adrian highlighted interoperability as the challenge he would like to solve, considering that “a basic framework to work with” is something that organisations need.
“We need to have a record that is stood up somewhere. We used to work with a product called Health Vault, from Microsoft, and it was a really good architecture, because the records say in the cloud, and it had a set of open interfaces that you could use. You could send information to it, and the patient could then link that to apps and other organisations. We’ve got the NHS login, which is brilliant, and facilitative for this kind of architecture, because patients can then move between different apps. We need to create an infrastructure where the different vendors and different functions can play with it.”
On what that might look like, Adrian focused on the need for “a standard for the record”, which “could be open EHR-based”. A security mechanism and a way of controlling which apps have access to the record would also be important, he continued, “so we need some governance, we need some regulation, we need a basic framework that people can use and then they can plug into”. “Really, what we’re looking for is a bit of leadership somewhere, and a bit of strategy”, he concluded.
Ensuring that a solution has the potential for “continuous improvement” was a challenge Manzoor selected for short-term benefit.
“So we don’t just stick with the solution we have, but its contents are evolving and adapting to the environment, so it can embrace new ways of working, new technologies, or whatever comes in the future.”
Gege talked about the need to understand the role of technology as an enabler, in enhancing clinician-patient relationships, saying: “For us, that means to design AI systems that support clinicians, that increase patient engagement, that handle routine enquiries and administrative tasks, with the benefit of freeing up more time for direct patient care. We’re living in a digital world, and it’s our responsibility as leaders to think of where the healthcare industry is going to be in twenty years, because the decisions in design and technology implementation we take now, will critically impact the future of healthcare delivery.”
Other challenges brought to the discussion by our live audience included a lack of information and education for the patient population; the lack of an environment which would “elevate adoption and trust towards the emerging digital health economy”; the need for “a cultural shift towards patient participation”; and the range of clinical systems resulting in “possible duplication of data input which is in no-one’s best interests”.
To find out more about EBO.ai ’s Intelligent Patient Portal, please click here or download the guide.
We’d like to thank all of our panellists for their time in sharing their insights and experience with us.