A recent survey by the UK’s Ada Lovelace Institute concluded that while the National Health Scheme (NHS) “is rattling toward a digital future in pursuit of efficiency and innovation”, there is:

“a fundamental difference in perspective between those working with data to plan healthcare interventions and design digital health services and those interacting with digital health services as patients. It exposes a misalignment between priorities for policy and industry regarding digital health services and what economically disadvantaged people (often with chronic or complex health issues) need and want from healthcare.”

So what’s going wrong?

While there certainly are problems with the design of individual digital health services, the Lovelace survey found that biggest problem was a collective failure of developers and administrators to recognise that “digital health services must be designed and deployed with the specific constraints of the social and political landscape in mind”. Lovelace identified two key overarching constraints: the digital divide and the lack of public trust in the health system.

The digital divide is bad for your health

Lovelace commented that, in our increasingly digital-first world, digital determinants of health are interacting with and complicating the traditional social determinants of health.

In England, 27 percent of the population, over 14 million people, have the lowest digital capability, without regular access to a device or the skills and confidence to turn on a device, use an app, log in and/or enter information on a digital device by themselves. In turn, about 30% of people who have no online access or use find the NHS to be one of the most difficult organisations to interact with.

There have been attempts in the UK to facilitate access to digital health services by providing users with dedicated terminals or digital devices which are restricted (the old strategy of “walled gardens”) to health services. However, a study by the King’s Fund found that working backwards from the health services to bridge the digital divide is ineffective:

“We heard from organisations and teams providing devices that they had felt inclined to restrict how the devices could be used, perhaps limiting them for only health purposes. However, this was found to be counter-productive. Restricting device functionality typically makes a device less useful and potentially less valued by individuals and so increases the likelihood of it being damaged, lost or not used. It’s also important to consider that non-health care activities, such as streaming music and using social networks, improve digital skills and digital confidence.”

Are we just digitalising a lack of trust in the health system?

While digital health services are often promoted as increasing access and improving useability in the health system, the Lovelace survey found a much deeper cynicism amongst health practitioners and users:

“Developers and commissioners of these systems must consider how their systems will be perceived at a time when the health of the UK population is declining and healthcare workers are striking on an unprecedented scale over poor pay and benefits. This poses a risk to patient trust: public perceptions and experiences of an underfunded, understaffed NHS can mean efforts to digitise health services are seen as cost-cutting first and foremost, rather than efforts to improve care.”

This cynicism about the ‘true’ motivations behind digitalisation of the NHS were deepened by perceptions that:

  • at the macro level, “there was a feeling of not having a say in how digital transformation is changing healthcare: …many felt like digital health services were being pushed through over their heads, with little agency or consultation in the matter”.
  • at the level of use of individual services, “there’s a question of whether there’s an actual choice to opt in or out, or if it’s ‘just the way it’s done now’, without accompanying advice and reassurance about how they work.”

The Lovelace survey observed that these sentiments about digital health services were “particularly salient in the aftermath of pandemic lockdowns.”

‘Computer says no’

The Lovelace survey found that even if people were willing to use the tech, digital health services were often inflexible or inadequate in their design. Digital health services were not seen as dealing well with complex medical problems. Designers – with their tidy, structured approaches to data fields – tend to default to a ‘one size fits all’ interface which defeat or overwhelm users with complicated or multiple problems:

‘It’s all right ticking boxes [but] sometimes you don’t feel that box is you, although it says ‘other.’ Other is just a word, it doesn’t mean anything.’

The ‘reductionism’ which digital apps can drive also means that broader social and personal factors in a person’s life which are important to diagnosis and treatment are missed, such as socio-economic factors. Participants reported that poor use of data adds a burden of having to explain themselves over and over when seeing different medical professionals.

Where digital health services are used to perform a screening or ‘triage’ role, they can leave the user with a sense of being ‘judged’:

‘I feel like it’s really unhelpful to be constantly reminded that I’m overweight and almost diabetic, but there not be any follow-up support for that ... It’s like, ‘This is a ‘you’ problem, so go and fix it. Okay, see you.’ Whereas what they don’t see, because it’s not on my records because I never went through services for my addiction. So what they don’t see is that I put all of this weight on because I was so busy focusing on trying to stay clean that I let other stuff go.’

The Lovelace survey also found that “[f]ar from a ‘digital front door,’ digital routes to the health service acted more like a brick wall for many people we heard from”:

‘One of the services that they have at the surgery is called e-consult whereby you get a form and you’re asked questions about the condition that’s troubling you. I did that twice when I had a problem … and at the end of it, on each occasion it said, ‘we can’t help you, you need to see a doctor’ – it just tells you to see a doctor and it’s because you can’t get to see a doctor that you did the e-consult in the first place. It’s a vicious circle.’

The Lovelace survey concluded that many digital health services failed the most basic and fundamental test: “the broader topic of digital healthcare kept coming back to this key issue of not being able to get an appointment or communicate properly with a healthcare professional (including GPs, specialists, and mental healthcare providers).”

A doctor in the loop

There is a sense from the Lovelace survey that developers and administrators are too ambitious in what they believe can be achieved through digital health services:

‘To avoid ‘techno-solutionism’ in response to the challenges that the NHS faces, it is important in commissioning and development to justify why a new digital health service is the best solution to the problem that needs solving, that is, that a digital solution is not implemented just because it is new and exciting.’

Respondents reported situations in which use of digital tools, such as remote video appointments, lead to misdiagnosis which might have been avoided with face-to-face contact with a health professional:

‘In one scenario I can remember, J’s doctor, he had an issue with skin, these things on his hands, and they actually asked me to take pictures of it and send it to them on WhatsApp. And I’m like, ‘What? Why do you just not want to see him?’ And it was, ‘Oh, we can deal with it this way … but she misdiagnosed his hands through the pictures, but you could just tell it wasn’t scabies, because I was touching his hands and I wasn’t getting it’.

But there is a broader sense in which digital health services are making health care less personal:

‘I think they’re already becoming far too reliant on telephone conversations, consultations and Zoom things, and yes, that’s not healthcare. That’s a very good buffer to see maybe who needs to come in sometimes but this isn’t healthcare.’

This could be dismissed as a Luddite reaction. But our health is probably one of the most personal things to us all. This impact of the shift to digitalisation can fall more heavily on the disadvantaged:

‘Many interview participants – especially those with a higher health burden, complex needs, or who may be ‘digitally excluded’– felt that digital health services are making care less personal, and they missed having an ongoing relationship with a clinician who knows them and their history.’

For some of the respondents, having to navigate digital health services made existing access issues (like long waiting times) worse by as patients felt even further from having someone to talk to or knowing the right pathway to take.

The Lovelace survey concluded that there’s a need for to engage more closely with individuals from different backgrounds and marginalised groups to understand their experiences and design services that meet their needs:

“Data collected about people may not reflect the nuances of their lived experience, and the impact of this gap may be more significant for those with higher health burdens, or who feel less agency in the healthcare system…..There is a real tension between a desire to better join up personal health data so clinicians get a good idea of what the patient is going through, and resisting over-datafication of patients or relying only on clinical data to understand someone’s situation.”

But that does not necessarily mean that the answer is to build a ‘better digital mousetrap’.

‘Digital health services that address these fundamental structural concerns and reduce admin and operational burden for clinicians may be more important now than shiny new tools that are parachuted into an under-resourced system.’

Sometimes there is no substitute for a human health professional, as rushed and pressured as they may be, to understand and treat a human patient.

Read more: Access denied? Socioeconomic inequalities in digital health services

Expertise Area