I’m trying a different format for weeknotes to see if I can be more consistent in keeping up with them if I limit myself to a few simple questions. Thanks to Sam Villis for the definitive survey of weeknote styles which I’ve used to choose this format. If I’m doing it right, the questions will vary week to week. Here goes…
1. What inspired me this week?
Meeting the shortlisted candidates for our NHS Digital User Centred Design Graduate Programme, and getting to spend time with my brilliant colleagues who volunteered to help out at the assessment centre. Growing the talent of the future has to be time well spent.
Team photo of all the user researchers, designers, content people, and graduates who helped in the assessment centre for our first ever cohort of NHS Digital user-centred design graduates
Meeting design master’s students at the Royal College of Art. They asked many smart questions about designing for health and care.
2. What connections did I make?
Spent a couple of hours in the user research lab with the National Data Opt Out and NHS App teams. For several of our teams, 2018 was a year of minimum viable product to get the basics right, one service at a time. Sitting in the observation room with members of both teams together, I understood how 2019 will be the year we connect all that stuff into coherent user experiences that work together to fit our emergent understanding of users’ goals.
3. What capabilities did I build?
I joined designers, developers, and testers in a session of accessibility training with Alistair Campbell of Nomensa. Alistair had tailored the training based on manual and automated testing of the NHS website, and used examples from the site to show us where things could be improved.
Conducted second interviews for a lead researcher on the NHS website. Exciting news on that soon.
4. What goals were set?
I uploaded my learning contract for the NHS Leadership Academy Nye Bevan Programme. The learning contract describes how I will demonstrate to myself and my peers in my learning set that I have met the programme learning objectives. I also ran through this with my director, Ian, who was encouraging and made helpful suggestions.
Picked up a high priority action to help one of our teams with their forthcoming service standard assessment. I was pleased we were able to pull together trained assessors at very short notice for a mock assessment on Monday.
5. What do I need to take care of?
One of our teams is feeling under pressure because a few people have left and we’re still in the process of replacing them. The work to bring in replacements takes up the team’s time too, so colleagues need to go easy on them in the meantime.
With all those brilliant, high-value things this week, I got to Friday having hardly made a dent in my email inbox. Around mid-week I took a hard look at the gaps in my calendar and identified the things I realistically had a chance of getting done by the end of next week. Note to keep prioritising, delegating and saying no to lower value work.
Messy desk with sticky notes, scissors, coffee cup and laptops
1. Something changed
On 5 July 2018, the National Health Service turned 70. I was exactly 13 months into my own time with the service.
According to my calendar, on that day I: met Lisa from GDS, attended the NHS website fortnightly show and tell, chaired the design governance board, took my cat to the vet, and heard a playback of user research on clinicians’ attitudes to the website.
The morning after the 70th anniversary day, I tweeted 3 truths as I saw them about innovation in the NHS. They’re in tension, but not contradictory…
The service as a whole needs more money. Innovation must be led by the people who deliver the service, the ones who know where real opportunities lie. They need some slack in the system to learn new skills and try new approaches.
The opportunities afforded by new technologies are massive and awesome. NHS people need to get smarter at spotting what’s genuinely transformational. Otherwise we’ll just squander any extra money on fads and snake oil.
The system is still rife with failure waste, much of it caused by poorly designed IT. The fixes for this are well known, systematic, and start with better understanding of human behaviours, not with force-fitting unproven new technologies.
The NHS belongs to the people. Those of us on the inside have a duty to be honest and truthful in carrying out our jobs. (That’s in the NHS Constitution.)
We also depend on critical friends outside NHS organisations. For our friends to contribute, we must set high standards of openness, assume criticisms are well-intentioned, and amplify the voices of people who don’t have privileged insider status.
At that point, we were in full on “strategy is delivery” mode, heading towards the Health and Care Innovation Expo, a massive annual conference-cum-trade-show of all things health and care. When teams are continuously delivering value to users, synthetic milestones like this ought not to matter much, but in our world they do, and even more so this year, on which more later.
There were pre-launch nerves, but we made it to the Expo in the first week of September with a visibly different, mobile-first, more accessible and usable NHS website — and a new name to top it off. No more “NHS Choices”. There’s still much more about the website that we can improve, but that was an important turning point for everyone on the team.
In another building half a mile across Leeds, the equally talented team on the NHS app were working independently to their own private beta timelines. Dean, our lead designer, became the shuttle diplomat making sure that app and website would look and feel part of the same family.
Also in time for Expo, we released a beta version of a new NHS digital service manual, including design principles created in collaboration with some of our partners and other NHS organisations. The design principles are in beta too, and we welcome your feedback.
Expo 2018 was a 2-day event in Manchester. I didn’t make it in person to the first day, but did catch some of the highlights streamed by our communications team colleagues.
I popped across the Pennines on day 2, mainly to support Rochelle and Tero, who were working with Simon and Kassandra from Public Health England on a session about user-centred design – which all four of them ran brilliantly.
At 3:30pm, the exhibition stands emptied out as everyone crammed into the auditorium for the main event, the Secretary of State’s keynote speech. For a flavour, try this sample quote:
“Mandated standards are just the start. Too often people with too little technical understanding are buying IT from suppliers who want to capture the buyer so they can’t ever go elsewhere. Suppliers’ interests are too often not aligned with the NHS’s interests, and the contracts badly managed. This supplier capture is common in IT, but it is not inevitable. So we are seriously going to increase the in-house capacity to understand the technology, to procure the right things, to manage them better, to split up big contracts into smaller pieces, to ensure an agile, iterative approach focussed on the question: what is the user need?”
Two days later I was back in Manchester, this time in a t-shirt not a suit, for UK Healthcamp. It was my first Healthcamp. What a lovely group of people to spend a Saturday with any time. In the light of Matt Hancock’s speech, and with his tech advisor Hadley in the room, it felt as if this band of health tech radicals’ moment might have arrived.
I learned loads at Healthcamp, put faces to names I’d long followed from afar, and posed a question of my own to a windowless basement room full of thoughtful healthcampers: “What do people need to be able to trust a digital health service?” It’s a question I’d been thinking about a lot, because the fifth of our design principles is “Design for trust”.
So this 6-month note breaks almost exactly into two distinct halves: before and after 6 September.
In the weeks that followed, while the NHS website team enjoyed a well-deserved boost in confidence and credibility, I was privileged to contribute in a few tiny ways to a set of vision and planning documents that make the new direction clear.
This work doesn’t always happen in the open, which is shame. We should constantly be asking “who is not in the room, but should be?” Nevertheless I’ve been hugely impressed by the commitment to user-centricity among the people tasked with making those plans, and by their openness to challenge and new ideas. As Ben Goldacre put it:
There’s a trolley, trundling through Whitehall and the NHS, with ‘Make NHS IT Better’ written on the side. If you throw your thoughts into that trolley, right now, they will join all the other ideas, problems and plans.
What purpose do these documents serve? Policy papers won’t fix the problem; people and delivery will. But I feel — really feel — that the last 3 months have made us all bolder. It has been lovely to see the emergence of a new coalition of people from across health and care rallying round the principles in the vision paper.
Among the people reacting, I see some archetypes. These are not a comprehensive taxonomy. I have been all of them myself at different times in my career.
Some people have long worked for a focus on user needs, privacy and security, interoperability and openness, and for inclusion. They’ve done this in the face of legacy technology and commercial arrangements, complex organisational and delivery structures, and a risk-averse culture that failed to register any of those things as the real risks. I hope they’re not now too worn down to seize this opportunity.
Some have focused on particular topics, such as clinical informatics, patient and public involvement, data geekery, privacy activism, agile procurement, or user-centred design. Often they’ve been stymied by the “dark matter” that arises from the other problems not visible inside their own paradigms. This vision demands a new, shared understanding between the different practices. None of us has all the answers.
Then there are the shy reformers, secretly inspired by the progress of user-centred, agile methods elsewhere in the public sector. They’ve been tamping down their enthusiasm, reining in their ambition, in order to make progress in legacy organisations. After reading the tech vision, they shouldn’t have to do that any more.
I’m mindful that others are still reserving judgement.
Some will look at the architectural principles in the vision and find a dozen reasons why their particular corner of the digital world is special. Telling them they’re wrong or stupid won’t help. We must show them respect, evidence, and examples, not just words.
And many have been burned before. After years of mismatched understanding between users and suppliers of healthcare technology, why should they believe that this time is different? The burden is on us to prove it is.
Service design, as a people-centred, holistic, co-creating discipline, could have a big part to play in bringing these tribes together around a shared purpose of better care and improved health outcomes for everyone in England.
2. Freedom to act
In my 12-month report, I mentioned the influence of David Marquet’s ‘Turn the Ship Around’. His over-riding imperative is to give control to the people who know best: to “move the authority to where the information is”. In order to succeed at this, he says, there are two pre-conditions: clarity and competence:
… as control is divested, both technical competence and organisational clarity need to be strengthened.
So if, on 6 September, or maybe on 17 October, the NHS reached a moment of digital clarity, how are we building the competence to make it real in the profession group I look after, design?
Whenever I need a reality check, I remind myself that our entry level design assistant role is on the same NHS pay band as an A&E paramedic; our lead designers are on a level with consultant psychologists. To make sure there’s a good reason for that, we’ve been though the NHS Agenda for Change job evaluation scheme, which assesses the knowledge, responsibility, skills and effort needed to do each job.
Technical skills and familiarity with design methods are part of the picture. But we’ve also had to explain what designers do according to a controlled set of generic responsibilities. I found that a more useful exercise than I expected. Here are some highlights against the Agenda for Change responsibilities:
Design is Research & Development – we discover needs. We make alphas and betas. Design is surely the silent third letter between R&D!
Design is Policy and Service Improvement – we take responsibility for the design of a new service, or substantial improvement to an existing one. We can show how different hypotheses for policy change are linked (or not) to user needs
Design is Quality – we champion key “non-functionals”, such as usability, accessibility, and desirability
Design is Equality, Diversity and Rights – we’re advocates for the principles of inclusive design
Design is Communication and Relationships – not only making clear and engaging visual artefacts, but also getting to know the people we need to work with to help make designs a reality
Design is Planning and Organisation – the most junior designer must be able to manage their own workload, and as they get more senior they take on greater strategic responsibilities, and organise others to deliver their designs
Design is Analysis and Judgement – knowing what techniques to use, and when, even with conflicting sources of information. The more senior the designer, the more this judgement comes to the fore.
There’s another factor that affects the NHS pay banding, called Freedom to Act. Our job descriptions set out the level of empowerment that each level of designer is expected to have in order for the organisation to get best value from their skills, knowledge and experience.
For example:
Design assistant (band 6) is directly responsible for the design outcome of their work
Associate designer (band 7) plans their own work within the remit of their assignment within the product team
Designer (band 8a) knows when to apply the relevant policies and guidelines and when to deviate from them
Senior designer (band 8b) acts as the final arbiter for design decisions within their product teams
Lead designer (8c) works autonomously, often in areas of novel practice, seeking advice from peers and industry experts outside the organisation.
Designers in a large organisation can easily be disempowered. Sometimes their managers or teammates don’t give them their freedom to act. Sometimes, it turns out that they’ve had the freedom but didn’t realise it, or were afraid to use it. Either way, we have a problem, and we’re wasting public money. In the coming months, I’m planning to spend more time talking with designers and their managers about freedom to act.
The senior designers have taken on the task of organising our quarterly team events – the last one in London, and the next one in Leeds. While we spend most of our time embedded in the different directorates and teams, it’s important to get together as a community of designers across the whole organisation.
Giving control doesn’t mean giving up on governance. The service manual, with components and patterns alongside it, is there to speed up delivery. We found we needed some lightweight governance to make decisions, when they’re needed, about overall design quality, interconnected design work from different teams, and what goes into the manual. So we’ve set up a design governance board, bringing together the right people on an ad hoc basis to review work that could be of value to more than one team.
It’s one of a number of ways, alongside the weekly design huddles, that we maintain cross-team visibility of what others are doing. Having these forums helps me trust teams to get on with their work, and rely on them to bring things in for peer review or expert guidance when they need it.
Standards and design governance help us to design things right. Service design takes the lead when it comes to making sure we design the right things. We’re still finding our way with service design, but we now have practitioners embedded in 4 different directorates or sub-directorates across the NHS Digital organisation chart. Service designers add the greatest value when they have licence to work broadly across a portfolio, such as Urgent an Emergency Care. I’m sorry that we haven’t been able to create this space for all the service designers we have brought into the organisation.
Over the last 6 months, I’ve mostly managed to keep my commitment to spending at least 2 hours every six weeks observing primary user research. There’s always something surprising and worthwhile to see and hear. We’re lucky to have Rochelle and Gemma and their team of user researchers.
One of the highlights of the past few months has been welcoming a new cohort of Digital Service Delivery graduates on a 2-year programme in which they’ll get to try out working in user research, design, content design, product and delivery management. For next year, we’re recruiting a first ever cohort of user-centred design graduates, who will focus on user research, design and content specifically. Our CEO, Sarah, has been a champion for apprenticeships and graduate trainees, and thanks to Amanda, our head of profession, we now have the strength in depth to be able to look after a larger number of trainees.
I’ve tried to make time to keep an eye on the future too. Thanks to Cassie for inviting me to Nesta’s event on “collective intelligence” in September. I found it a timely counterpoint to the tech determinist narratives around “artificial intelligence”. This in turn shaped the talk I gave at Interact 2018. (Here’s the video and my write-up of that.) I think the other speakers and I were mostly in violent agreement. For the Interact talk, I also drew on the principles in the initial code of conduct for data-driven health and care technology. A new version of that is due to be published soon.
I was also honoured to be invited to talk at Health Product People at the Department of Health and Social Care, especially so because I was on the same bill as Hadley, Ian and Kassandra, all of whom were brilliant. I loosely titled my talk “Stop disempowering people.”
3. What’s next
I still wake up every day with a sense of wonder that I get to be head of design at NHS Digital. It’s a massive privilege. I’m also in awe of my talented colleagues, who could no doubt command gigs in many more glamorous brands and sectors, but choose to work on stuff that matters. I hope never to take any of them for granted.
Meanwhile things are changing in the organisation. We have some new leaders. The matrix management structure, of which I was never a massive fan, is being dismantled. A simpler, clearer, and hopefully faster, directorate structure is taking its place. Along with other senior managers in NHS Digital, I’ve had to reapply for my job. (Good news: I passed!)
At times during this transition I have felt an odd mix of responsibility without power and power without responsibility.
Responsibility without power, because I still feel responsible for the team I was hired to lead, but in the new structure, I have lost some of the levers I previously had to help them.
Power without responsibility, because, thanks to the tech vision, insights on user-centred design and digital are more in demand than ever across the wider health and care sector, but the programme and line management structures of my own organisation are still on a journey to reflecting that.
Amanda, the head of profession who hired me, is now leaving to take up a digital director role in the charity sector. I will always be grateful to have had Amanda as my manager here. Uniquely, she empowered everyone who worked for her, making the space for me and the other deputy heads of profession to grow in confidence and expand our teams. I’ve been given the opportunity to cover as interim head of profession, looking after product managers, delivery managers, user researchers and content people as well as designers. This I will do with a sticky note on the inside cover of my notebook that reads: “What Would Amanda Do?”
I always said this was a multi-year commitment, and as I head into the second half of year two, this job is not done yet, not by a long chalk. In the next 6 months, alongside working on the most pressing and worthwhile transformation portfolio in the world right now, I hope to:
Challenge more on design quality and performance across all our products and services
Kick off some work on user experience strategy across the wider health and care system. User experience is not the whole answer, but it’s a good place to start understanding Why Doctors Hate Their Computers
Further raise the profile of design and the other digital profession roles in the NHS, including a keynote at Service Design in Government 2019
Look after my own long-term professional development, by applying for the Nye Bevan Programme of the NHS Leadership Academy.
Text on screen behind speaker: “Stop disempowering people.”
I was honoured to be invited to talk at Health Product People at the Department of Health and Social Care this week, especially so because I was on the same bill as Hadley, Ian and Kassandra, all of whom were brilliant. I loosely titled my talk “Stop disempowering people.” (Thanks to Ian for photo of me in full flow.) Here’s roughly what I said…
Text on back of phone: “What is the user need?”
This is former UK Government Chief Technology Officer Liam Maxwell’s phone. On the back he famously had the words, “What is the user need?” When a meeting seemed to be going off track, down some rabbit hole of business sophistry, Liam would show people the back of his phone. User needs are always a good place to start.
But then where? And which users, which needs?
I talked about how, in health and care, we need to consider people’s clinical, practical and emotional needs. (Thanks to Janet Hughes for framing that so pithily for us.)
I also laid down a challenge to the assembled product people: We talk a lot about empowering people. But who are we to empower anyone? Everyone starts with power; it’s our job to stop them being disempowered.
Illness, disability, social circumstances and fear can all be disempowering. So, unfortunately, can be the complexity of the health and care system, and the way we treat people within it.
Recently I heard the story of someone who had lived with anxiety, and had developed their own ways of coping with it. Things went downhill for them the day they were referred to a service with a long waiting list. While in that limbo, they were disempowered – they knew they needed professional help, but had no way of accessing it. The service had taken someone who was managing, and disempowered them, made things worse before making them better.
Digital service, done right, can help people to recover their power over their own health and care.
I gave the example of a new section on the NHS website about Type 1 Diabetes. Our team identified that the point where people most needed help was in the weeks and months after getting a new diagnosis. Instead of structuring information around a clinical taxonomy, or around the way our services are organised, they created a section just for people who were newly diagnosed. By bringing together some key information in a simple package, they could help people take charge of their own health and care at a time when there’s a chance of being overwhelmed.
I might have subtitled the talk “People showing their phones”, because in addition to Liam’s phone, I also showed this picture…
A health and care worker pointing at her very old, battered, dumbphone
It’s from research I did before I joined NHS Digital, as part of a discovery for Leeds City Council and m-Habitat [PDF link – sorry]. We were investigating how health and care professionals could increase their digital capability. But often, the barriers turned out to be more basic than capability. Research participants would show us the battered dumbphones they’d been given to do their jobs. Fix this first, they’d say, then we’ll be interested in the digital tools you have to offer.
We met social workers who were trusted to go into difficult situations, and make decisions with massive implications for families’ futures. Yet they hadn’t been trusted to connect to the wifi at a local health centre. (The situation in Leeds has, I believe, got much better since then!)
It became clear to our whole project team that if we didn’t empower our workforce, they could never empower patients or service users.
Our new design principles are intended to empower. They don’t tell people how to do their jobs – but rather provide a set of principles within which anyone can apply their own expertise, and their understanding of the situation at hand.
If we’re serious about putting people in control, we need first to free our minds of deficit thinking. I presented four manoeuvres that could help to do that.
1. User needs assets
The first, and perhaps most obvious, is to mitigate the deficit thinking implicit in the term “user needs” by remembering also to consider users’ assets – what they have and can do, not what they lack or need fixing.
Assets are a design material, something we can work with to help people achieve the outcomes they seek.
For example, a user might need to exercise more, but their love of music is the asset that suggests dance classes as a good way to get started.
2. Capability stories
Who is the hero of the user stories you write? Make sure it’s the user by focusing on the last part of the construct, the bit that starts “so I can…”
Let’s call it the capability story, something that doesn’t just solve short-term problem, but leaves the user stronger and more able than before.
For example:
“As a person who has just been diagnosed with type 1 diabetes, I need to learn how to count carbs, so I can control my blood glucose levels.”
“As a person with memory problems, I need information that’s easy to understand, so I can decide for myself.”
Or maybe even:
“As a visually impaired web user, I have a screen magnifier on my phone so I can read web pages, provided they’re well designed and coded.”
3. Ladder of participation
Sherry Arnstein‘s ladder of citizen participation posits that there are different levels of involvement, from manipulation, through placation, and rising up through partnership to true citizen control.
Nobody wants to feel manipulated, and when they do, it’s unlikely to achieve the manipulator’s intent.
In digital service, the most passive level is being treated as a data subject – a bundle of behaviour to be analysed and influenced.
We try to avoid casting the people who help us with user research as lab rats – not testing the user, but testing with the user, taking part in research.
But let’s not stop there. I was on a team at Co-op Funeralcare with four funeral directors who were taken off funeral duties for 6 months to help the digital team design new tools and workflows.
In health and care, we should consider patient leaders as our most esteemed stakeholders, invite them to our sprint reviews, and give them seats on our programme boards.
I wanted you to see that film – 70 years old this year – because its promise, of universal healthcare, free at the point of need, is as relevant today as in 1948. But much has changed in that time. 97% of people born in Britain since 5th July 1948 – myself included – have been born in the NHS. The cartoon baby at the end of that film is now himself a cartoon pensioner. He is living a longer, healthier life than his cartoon parents. Meanwhile medicine is transformed. We’re on the cusp of another revolution in genomics and personalised medicine.
I work for an organisation called NHS Digital. We’re the national provider of data and technology services for health and social care.
Our users are:
53 million patients and the public across England
1.3 million health and care professionals
We support:
the scientific research community who use NHS-scale data for research, and
the service community – the people who work every day to keep the health and care service functioning, and to make it more efficient.
As head of design, I’m privileged to work with a great team of designers, user researchers and others who together are designing and delivering some vital tools and services in support of that vision. I want to tell you about that design capability. I want to show you some of the things we’re making. And I want to think about the future of health and care, and how design has a vital role to play in realising its promise.
Design capability
Let’s start with the design capability we’re building. Around 40 designers work on services used by patients and professionals across the complex health and care ecosystem. There are a few services – like the national NHS website – that we design and build as national solutions from the ground up. But for many others, the digital bit we deliver nationally is just a small part of someone else’s end-to-end service, including elements bought and built locally by NHS organisations or third-party suppliers.
The make-up of our design team reflects this:
23 interaction designers
10 service designers
5 graphic designers
We have interaction designers working directly on the digital services we deliver ourselves. They also need a graphic design sensibility, because we’re working with one of the UK’s most trusted brands, the blue lozenge. Above this, we’re building our service design capability, to be able to design health and care services, as Lou Downe at GDS says, from end-to-end and front-to-back.
It won’t come as a surprise to anyone who knows UK government design practices that our designers work in agile, multidisciplinary teams, along with product managers, user researchers, developers and other specialist roles.
Working at health service scale gives our designers another responsibility – to talk to each other – individuals and interactions over processes and tools. As the team grows, we’re also investing in design leadership capability – to have a mix of experience and seniority, a career path, and job descriptions. I might be the only person in the team who’s excited about this, but we recently got organisational sign-off for a complete set of designer job descriptions on the same NHS pay scales as nurses, pharmacists, and managers.
When I’m explaining what we do to non-designers – and, believe me, I spend a lot of time doing that – I always fall back on Jared Spool’s wonderfully economical definition of design as “the rendering of intent.” Intent without rendering gives us a strategy but cannot make it real. Rendering without intent may be fun – may even be fine art – but is, by definition, ineffective. Every one of our designers must be able to explain the intent behind their work.
Because in health and care, intent is varied and complex. User needs for health and care services come in at least three flavours:
There are clinical needs – people expect the services we offer, and the tools we recommend, to be clinically safe and effective. But that alone is not enough.
To be adopted and used, services must meet people’s practical needs in the context of their lives. If that context is mobile, the service must be designed mobile first. If it’s to be used in the middle of the night, it needs to connect to services that are open.
Finally, it’s as important to meet our users’ emotional needs. Sometimes people go to the doctor not just for information, but for reassurance. Information could be clinically accurate, but if it doesn’t connect emotionally, the user need has not been met.
Every designer in the NHS needs intelligence and empathy. They must understand the true intent behind their services, reconciling diverse clinical, practical and emotional needs.
Too easily, people become disempowered by worry, illness, disability, or social circumstances, and (even though we don’t mean to) by the way we have designed and delivered NHS services in the past.
So we need to design for the positive role of patients and carers, to think about their assets – what they have and can do already – as well as their needs and deficits. We need to co-design with patients, staff, family carers and voluntary sector who all make up our service community.
Finally we need a continual focus on digital inclusion and accessibility, because the NHS is for everyone, and those with the greatest health needs are also the most at risk of being left behind digitally.
This is Hastings, where our partners on the NHS Widening Digital Participation programme have been developing models of digital health interventions for people who are homeless or at risk of homelessness.
Show the things
For a long time, we had this quote on the wall by the team working on the NHS website redesign:
“It’s just a website, We’re not going to the moon.” Mikey Dickerson on fixing healthcare.gov
This summer, we transitioned the national NHS website to a new mobile-first, accessible platform. In 2018, this should be basic stuff really, but through a series of policy twists and turns, getting to this point was a big deal. We topped it off with a new name. In research, we asked people what they called it. And we used the words they used. So the site formerly known as NHS Choices is now simply “The NHS website.”
It’s just a website – with more than 48 million visits per month. That’s a quarter of all health-related web traffic in the UK. People expect a different relationship with healthcare, as well as different channels to access it.
There’s good evidence that people in control of their own health and care get better health outcomes. So while we’re here as the NHS for you in the times of greatest need, we also want to help you look after your own health and wellbeing even when you’re feeling fine.
The NHS website has to be there for users in many contexts, needs and emotional states. We have to design for the end-to-end user journey – whether a short, acute, episode of care, or management of a long-term health condition. Making the whole journey visible to everyone involved is powerful, because otherwise no one professional or organisation ever sees the whole picture.
One team looked deeply at the experience of people with Type 1 Diabetes. They mapped a whole journey from someone not even knowing they have a problem, through the trigger that leads to diagnosis, getting to grips with a potentially lifelong condition, and over time confidently managing their wellbeing. They found – and this has echoes in other conditions too – that the point at which the NHS website could help most is the first weeks and months after diagnosis.
Another team has taken the triage pathways that underpin the 111 non-emergency phone service and turned them into an online service. The online service is designed to get people to the help they need while taking pressure off the telephone service. But the big picture here is important, if we get it wrong, we could send worried well people in greater numbers to accident and emergency departments.
Confidentiality, trust and consent are big, complex issues we have to navigate. We have to understand them when delivering a simple, secure way for people to log in to NHS services. And we’re giving every patient control over how their data can be used beyond their own direct care.
Design doesn’t stop at the big picture. We also have to care about the details. Here, for example, is a page on the old NHS Choices website about paracetamol. The information on the old website was clinically safe and accurate. You’d expect nothing less of the NHS. But, in research, one group in particular – parents with young children – told us the way it mixed information about adults and children was disconcerting. They worried about accidentally giving an adult dose to their child. So we’re splitting the page in two – one about paracetamol for children, another about paracetamol for adults. That way we can meet not only the clinical need, but our users’ practical and emotional needs as well.
We had to diverge before we could converge on a single set of styles. When I joined the team a bit more than a year ago, everyone was telling me we had to deal with the inconsistent styles that were springing up in different teams. But I worried we were in danger of getting stuck too soon at a “local maximum”, super-optimising the first designs the NHS alpha team had come up with. So I made myself unpopular by telling designers to spend a bit longer solving their own problems, designing solutions in response to user needs.
When we did come to converge, through a process of patient design diplomacy by Dean, our lead designer, I believe the results were stronger for this extra round of divergence.
While we’re getting our own house in order, developing a consistent set of styles and design principles for the nationally-delivered NHS services, patients the public, and professionals experience a patchwork of interactions commissioned and delivered in many different ways across the wider NHS family. We want to make experiences consistent, no matter whether you’re using the national website, or a condition-specific app, or a service built by one of your local NHS organisations.
So we’re publishing our user-centred design standards, patterns and practices in a new NHS digital service manual. We’re working across the system to do this, and combining good public sector practice, such as the GDS design principles, with the things that make the NHS unique. The service manual team had hoped for a soft launch, but the beta has already been well-received, and I reckon we’re going to speed up this work in the near future.
Aside from the NHS website, we’ve had a team building a beta version of a new NHS app.
Design for the future
What role might human-centred design play in realising the promise of new medicine and technology? Look at the amazing trajectory of human understanding of DNA, RNA, enzymes, proteins, the genome, and the mechanisms by which they interact. This stuff will transform – is already transforming – our relationships with medicine. Crucially this generation of scientists are looking inside a black box, where their predecessors could observe its effects but not its inner workings.
At the same time, fuelled by petabytes of readily available data to digest, computer science risks going the other way in the framing of artificial intelligences: moving from explicable, simple systems to ones where it’s allowed to say, “this stuff is so complex that we don’t know how it works. You have to take it on trust.”
When we apply artificial intelligence (AI) to healthcare, transparency is essential; black boxes must be considered harmful. It’s not just me saying this. Here are the words of the Institute of Electrical and Electronics Engineers (IEEE):
“Software engineers should employ black-box software services or components only with extraordinary caution and ethical care, as they tend to produce results that cannot be fully inspected, validated or justified by ordinary means, and thus increase the risk of undetected or unforeseen errors, biases and harms.” — Ethics of Autonomous & Intelligent Systems
Transparency must be the order of the day. It comes in (at least) two flavours: the first is clear intent; the second, understandable operation. Both are under threat, and designers have a vital role to play in saving them.
When any technology moves from pure to applied science, intent must be centre stage. If we fixate too much on the computer science of AI, and not enough on the context of its application, intent will always be unintentionally obscured.
Many discussions about the “ethics” of AI or genomics are really, I think, discussions about the opacity of intent. If we don’t know who’s setting the goals for the machine, or how those goals are derived, how can we know if the intent is good or bad? For health and care, we have a new code of conduct intended to make sure this doesn’t happen.
In the words of Professor Margaret Boden, “the computer couldn’t care less.” She says:
“…computers don’t have goals of their own. The fact that a computer is following any goals at all can always be explained with reference to the goals of some human agent. (That’s why responsibility for the actions of AI systems lies with their users, manufacturers and/or retailers – not with the systems themselves.)” — Robot says: Whatever
It’s time for designers to double down on intent – true human intent that can be difficult to encode. In a domain as complex as health and care, intent is rarely straightforward. It can be changing, conflicting and challenging to untangle:
a boy was triaged on first contact as in less urgent need, but has suddenly taken a turn for the worse
an elderly woman wants to get home from hospital, but her doctors need first to be sure she’ll be safe there
the parents want to help their children lose weight, but know that pester power always leads them back to the burger chain.
User-centred design must clarify who the service is for, what problem they’re trying to solve, and what benefits we expect them to realise.
It’s time for designers to double down on intent, and – let’s be honest – this is not an area where design has always covered itself in glory. We know what design without intent looks like, right? It’s an endless scroll of screenshots presented without context – the Dribbblisation of design. If you think that was bad, just wait for the Dribbblisation of AI. Or the Dribbblisation of genomics.
Thoughtful designers on the other hand can bust their way out of any black box. Even if they’re only called in to work on a small part of a process, they make it their business to understand the situation holistically, from the user’s point of view, and that of the organisation.
Experienced designers are confident moving up and down the stack – through graphic design, interaction design and service design problem spaces. Should we point an AI agent at optimising the colour of the “book now” buttons? Or address the capacity bottlenecks in our systems that make appointments hard to find?
One of my team recently talked me through a massive service map they had on their wall. We discussed the complexity in the back-end processes, the push and pull of factors that affected the system. Then, pointing at a particular step of the process: “That’s the point where we could use machine learning, to help clinicians be confident they’re making a good recommendation.” Only by framing the whole service, could they narrow in on a goal that had value to users and could be usefully delegated to AI.
Designers are well placed to show the workings of their own (and others’) processes, in a way that proponents of black box AI never will. This is my second flavour of transparency, explainability, clarity of operation. Show what type of algorithm you are building, why that algorithm, how you check if it’s working the way you intended.
How might we:
communicate probabilities and uncertainties to help someone decide what to do about their disposition to a form of cancer?
show someone exactly how their personal data can be used in research to develop a new treatment?
involve people waiting for treatment in the co-design of a fair process for prioritisation?
In a world of risks and probabilities, not black and white answers, we should look for design patterns and affordances that support people’s understanding and help them take real, fully informed, control of the technologies on offer. This is not an optional extra. It’s a vital part of the bond of trust on which our public service depends.
The cultural ascendancy of AI poses both a threat and an opportunity to human-centred design. It moves computers into territory where designers should already be strong: exploration and iteration. I’m critically optimistic because many features of AI processes look uncannily like a repackaging of classic design technique. These are designerly machines.
Finding patterns in a mass of messy data?
Learning from experiments over many iterations?
Sifting competing options according to emerging heuristics?
User-centred design does all those things too.
Some diagrams explaining AI processes even resemble mangled re-imaginings of the divergent/convergent pattern in the Design Council’s famous double diamond. The threat is that black box AI methods are seen as a substitute for intentional design processes. I’ve heard it suggested that AI could be used to help people navigate a complex website. But if the site’s underlying information architecture is broken, then an intelligent agent will surely just learn the experience of being lost. (Repeat after me: “No AI until we’ve fixed the IA!”)
Designers should embrace the new, more design-like metaphors of rendering intent. As a profession, we have a great story to tell. We should talk more about our processes for discovering and framing problems, generating possible solutions and whittling them down with prototypes and iteration. Sure, we’ll need new skills, to change and evolve our methods – we’ve already mastered web, mobile, assistive tech. As Ursula le Guin wrote:
That’s the neat thing about technologies. They’re what we can learn to do.
As the title of Ellen Broad’s wonderful book has it, AI is ‘Made By Humans’. We can pair human intelligence with artificial intelligence, and harness the combined power of us all, through collective intelligence.
How might we give power to the communities of health and care (and help them understand each other better in the process)?
Patient community – taking advantage of collective knowledge and data
Clinical community – integrating machine learning into clinical practice
Scientific community – extending existing modes of collaboration
Service community – co-ordination problems and realtime system status
(Credit to Stefana Broadbent for framing the first three categories at a recent Nesta event on Collective Intelligence.)
Some people say that the pace of change in accelerating, and that big organisations like the NHS can never keep up. I don’t believe that. For 70 years, the NHS has known nothing but change. Back in 1948, Nye Bevan, the founder of the NHS said something remarkably prescient. I think this attitude is one of the reasons, against the odds, we’re still here as a 70-year-old institution today.
“We shall never have all we need. Expectations will always exceed capacity. The service must always be changing, growing and improving – it must always appear inadequate.” — Nye Bevan, 2 June 1948
A couple of Saturdays ago, still buzzing from a week of NHS website and service manual launches, and the NHS Expo, I took part in my first UK Health Camp.
I learned loads, put faces to names I’d long followed from afar, and posed a question of my own to a windowless basement room full of thoughtful healthcampers: “What do people need to be able to trust a digital health service?”
It’s a question I’ve been thinking about a lot, because the fifth of our new NHS design principles is “Design for trust”.
I ran the session as a loose variant of the 1-2-4-all liberating structure. Asking people to think about the question first as individuals, then in growing groups, the format was a great way of eliciting contributions from everyone in the room, then distilling down to some common themes.
At the end of the session, I left the room with a stack of sticky notes on which I had scribbled the key themes as groups reported back. Below is a summary with my own grouping and interpretation of the themes after the event.
The weighing of trust starts before we use a service, as we evaluate it to see if it’s going to meet our needs.
A few groups in the session talked about relevance: will it help me achieve what I need to do? To be relevant, a digital service will likely have to be part of an end-to-end journey, quite possibly including both digital and non-digital elements. Even in this digital world, having an offline presence is one of the things that can give a service credibility.
Once we believe a service might be useful, the next question, not far behind, is “has it been tested – for safety, practicality, and effectiveness?”
We trust things that come recommended by people we trust. Reputation matters, especially when expressed through peer recommendation. We make decisions about services in a web of relationships; a service will be more trusted if it is “culturally embedded”. In the context of British healthcare, there’s nothing more culturally embedded than the NHS.
To earn trust fully, there are things a service has to demonstrate in use.
Is it confidential? People set high standards for data protection, security and privacy. A service shouldn’t collect data it doesn’t need, and must be totally anonymous when you need it to be.
Is it personal? Provided confidentiality is assured, one of the ways a service can gain credibility is by showing information that only it should know. While anonymity is sometimes necessary, so too can be personalisation.
Is it transparent? Transparency of intent and clarity of operation are essential for any digital health service. Why is it asking me this? How did it get to that answer? Is it clear what I’m consenting to?
It is professional? The boring qualities of stability, reliability and consistency should not be underrated. If they go missing, trust in a service will be rapidly undermined.
Finally, there’s a quality of continuous improvement. Without this any trust gained is likely to be short-lived. Does the service take feedback? Is it accountable for its actions? Can you see in its present state the traces of past user feedback? “You said… we did…”
Those were the combined ideas of a self-selecting group one Saturday in Manchester. But tell me what you’d add? What would you need to be able to trust a digital health service?
On my holiday, I started reading into some topics I ought to know more about: artificial intelligence, genomics, healthcare, and the fast approaching intersection of the above. Here follow some half-baked reckons for your critical appraisal. Please tell me what’s worth digging into more. Also where I’m wrong and what I might be missing.
1. Opening the black box
By Yikrazuul CC BY-SA 3.0, from Wikimedia Commons
Reading Siddhartha Mukherjee’s ‘The Gene: An Intimate History’, I discovered the amazing trajectory of human understanding of DNA, RNA, enzymes, proteins, the genome, and the mechanisms by which they interact. There’s no doubt that this stuff will transform – is already transforming – our relationships with medicine. Crucially this generation of scientists are looking inside a black box, where their predecessors could observe its effects but not its inner workings.
At the same time, fuelled by petabytes of readily available data to digest, computer science risks going the other way in the framing of artificial intelligences: moving from explicable, simple systems to ones where it’s allowed to say, “this stuff is so complex that we don’t know how it works. You have to take it on trust.”
When we apply artificial intelligence (AI) to healthcare, transparency is essential; black boxes must be considered harmful.
It’s not just me saying this. Here are the words of the Institute of Electrical and Electronics Engineers (IEEE):
“Software engineers should employ black-box software services or components only with extraordinary caution and ethical care, as they tend to produce results that cannot be fully inspected, validated or justified by ordinary means, and thus increase the risk of undetected or unforeseen errors, biases and harms.” — Ethics of Autonomous & Intelligent Systems [PDF]
Transparency must be the order of the day. It comes in (at least) two flavours: the first is clear intent; the second, understandable operation. Both are under threat, and designers have a vital role to play in saving them.
2. The opacity of intent
It’s a commonplace to say that technology is not neutral. I won’t labour that point here because Sara Wachter-Boettcher, Ellen Broad and others do a good job of highlighting how bias becomes embedded, “AI-washed” into seemingly impartial algorithms. As the title of Ellen’s wonderful book has it, AI is ‘Made By Humans’.
That doesn’t seem to stop stock definitions from attempting to wall off AI beyond the purview human control:
“In computer science, AI research is defined as the study of ‘intelligent agents’: any device that perceives its environment and takes actions that maximise its chance of successfully achieving its goals.” — Wikipedia
But what goals exactly? And how did the AI get them? The Wikipedia definition is silent about how goals are set, because, in the words of Professor Margaret Boden, “the computer couldn’t care less.”
“…computers don’t have goals of their own. The fact that a computer is following any goals at all can always be explained with reference to the goals of some human agent. (That’s why responsibility for the actions of AI systems lies with their users, manufacturers and/or retailers – not with the systems themselves.)” — Robot says: Whatever
When any technology moves from pure to applied science, intent must be centre stage. If we fixate too much on the computer science of AI, and not enough on the context of its application, intent will always be unintentionally obscured.
Many discussions about the “ethics” of AI or genomics are really, I think, discussions about the opacity of intent. If we don’t know who’s setting the goals for the machine, or how those goals are derived, how can we know if the intent is good or bad?
Moreover, true human intent may be difficult to encode. In a domain as complex as health and care, intent is rarely straightforward. It can be changing, conflicting and challenging to untangle:
a boy was triaged on first contact as in less urgent need, but has suddenly taken a turn for the worse
an elderly woman wants to get home from hospital, but her doctors need first to be sure she’ll be safe there
the parents want to help their children lose weight, but know that pester power always leads them back to the burger chain.
In these situations, even Moore’s Law is no match for empathy, and actual human care.
3. Designers to the rescue
Design, in Jared Spool’s wonderfully economical definition, is “the rendering of intent.” Intent without rendering gives us a strategy but cannot make it real. Rendering without intent may be fun – may even be fine art – but is, by definition, ineffective.
It’s time for designers to double down on intent, and – let’s be honest – this is not an area where design has always covered itself in glory.
We know what design without intent looks like, right? It’s an endless scroll of screenshots presented without context – the Dribbblisation of design. If you think that was bad, just wait for the Dribbblisation of AI. Or the Dribbblisation of genomics. (“Check out my cool CRISPR hacks gallery, LOL!”)
Thoughtful designers on the other hand can bust their way out of any black box. Even if they’re only called in to work on a small part of a process, they make it their business to understand the situation holistically, from the user’s point of view, and that of the organisation.
Design comes in many specialisms, but experienced designers are confident moving up and down the stack – through graphic design, interaction design and service design problem spaces. Should we point an AI agent at optimising the colour of the “book now” buttons? Or address the capacity bottlenecks in our systems that make appointments hard to find?
One of my team recently talked me through a massive service map they had on their wall. We discussed the complexity in the back-end processes, the push and pull of factors that affected the system. Then, pointing at a particular step of the process: “That’s the point where we could use machine learning, to help clinicians be confident they’re making a good recommendation.” Only by framing the whole service, could they narrow in on a goal that had value to users and could be usefully delegated to AI.
4. How do you know? Show your thinking.
Crucially, designers are well placed to show the workings of their own (and others’) processes, in a way that proponents of black box AI never will.
This is my second flavour of transparency, clarity of operation.
How might we:
communicate probabilities and uncertainties to help someone decide what to do about their disposition to a form of cancer?
show someone exactly how their personal data can be used in research to develop a new treatment?
involve people waiting for treatment in the co-design of a fair process for prioritisation?
In a world of risks and probabilities, not black and white answers, we should look for design patterns and affordances that support people’s understanding and help them take real, fully informed, control of the technologies on offer.
This is not an optional extra. It’s a vital part of the bond of trust on which our public service depends.
5. Designerly machines
Applying fifty iterations of DeepDream, the network having been trained to perceive dogs – CC0 MartinThoma
The cultural ascendancy of AI poses both a threat and an opportunity to human-centred design. It moves computers into territory where designers should already be strong: exploration and iteration.
I’m critically optimistic because many features of AI processes look uncannily like a repackaging of classic design technique. These are designerly machines.
Dabbers ready, eyes down…
Finding patterns in a mass of messy data? Check!
Learning from experiments over many iterations? Check!
Sifting competing options according to emerging heuristics? House!
Some diagrams explaining AI processes even resemble mangled re-imaginings of the divergent/convergent pattern in the Design Council’s famous double diamond.
“A diagram outlining a forward pass though our three 3D generative systems.” – Improved Adversarial Systems for 3D Object Generation and Reconstruction [PDF]The threat is that black box AI methods are seen as a substitute for intentional design processes. I’ve heard it suggested that AI could be used to help people navigate a complex website. But if the site’s underlying information architecture is broken, then an intelligent agent will surely just learn the experience of being lost. (Repeat after me: “No AI until we’ve fixed the IA!”)
The opportunity is to pair the machines with designers in the service of better, faster, clearer, more human-centred exploration and iteration.
Increased chatter about AI will bring new more design-like metaphors of rendering that designers should embrace. We should talk more about our processes for discovering and framing problems, generating possible solutions and whittling them down with prototypes and iteration. As a profession, we have a great story to tell.
A resurgent interest in biology, evolution and inheritance might also open up space for conversations about how design solutions evolve in context. Genetic organism, intelligent software agent, or complex public service – we’re all entangled in sociotechnical systems now.
The chance to spend a day in the company of 500+ people who share the same simple, highly motivating principles.
A movement so nicely positioned; both inside and outside; #OfTheGovernment, not owned by the Government.
A show of strength in numbers. We filled the QE2 Centre! Now nobody need ever feel alone, even if they’re the first OneTeamGov participant from their particular agency or department.
Seeing a Liberating Structure in action to fill a 14 parallel track, 70 session unconference grid with minimal faff – amazing! James’ write up of the method is much too modest.
The hard but unobtrusive work of the brilliant OneTeamGov volunteers. Every detail of the participant experience considered, so that everyone else could be fully present and enjoy the day.
Seeing familiar faces again. A who’s who of reform in digital government, including a fair few who passed through the learning and development courses I helped to produce and deliver at GDS and DWP.
Meeting new colleagues and kindred spirits from around the world. In particular those who I’ve followed on Twitter from afar. Also, Canada wins at stickers.
The session I pitched had a deliberately ambiguous title – I wanted to know, when participants saw the word “people,” whether they thought first of the people we serve, or the people who deliver service? It turned out to be both, and a consensus that we can only empower citizens if staff are also empowered. It’s also my contention that it’s not the government’s job to empower anyone – everyone should start with power, our job is not to disempower them.
At the end of the session everyone came up with a thing they could do to stop disempowering people. Sticky notes were as follows:
Give people data to help them improve their work. Stop using it to performance manage.
Make decisions based in data and evidence
Set boundaries to empower people in a hierarchy
Consider the impact of small actions in the context of power
Ask your team to make decisions and state intent rather than ask permission
Use the right language. Team leader -> coach?
Key question: What do you think?
Inform. Inspire, Empower.
Emphasise how the different systems disempower citizens.
Empower citizens with familiar systems across borders
Share stories of success when people use their initiative
Use your power to amplify voice of those less powerful
Ask questions that people know the answer to
We need to cut out the time we ask people to spend interacting with government. What things can you do in your org to reduce the stress / time people have to spend
Managers need clear job descriptions out in the open
Redefine management roles – to not make decisions
Redefine the role of manager – coaches/enablers
Amplify voices of others + seek diversity of views
