Paper forms were costing the NHS trust. In more than just money.
Most NHS trusts still clerk patients on paper. Forms get lost, data gets mangled in translation, have you seen a doctor's handwriting, and somewhere downstream a clinical coder can't do their job properly and the trust doesn't get paid what it's owed. NHS Buckinghamshire Trust knew this had to change. What they needed wasn't someone to digitise the forms. It was someone who understood why the forms needed designing around the humans who used them.
The problem
Paper forms aren't just inconvenient. They're expensive.
Paper patient clerking forms get lost. They have no audit trail. Handwriting is unreliable. They have to be scanned in after the fact, which takes time and produces something that's stored but not always usable. And when the data is wrong or missing, clinical coders can't do their job properly, which means the trust doesn't get paid correctly.
There's a bigger picture too. The NHS has a long-term ambition to build a single patient record for every patient in England. That's not possible if a significant chunk of clinical information lives on paper, or in scanned images that no system can read.
The trust had bought a platform to solve this. What they needed was someone who could design the forms properly, so the platform had something worth building.
The complication
Clinical forms aren't just forms. They're logic.
A paper form gets away with being vague because a human fills it in and applies judgement on the fly. A digital form can't do that. Every field needs a decision made upfront: is it required or optional? And if a clinician records a particular finding, what does that unlock? What information is now needed that wasn't a moment ago? Get that logic wrong at the design stage and you're building the ambiguity in permanently. But get it right, and you've got something that saves time, money and lives.
The forms also needed to work for multiple audiences. Clinicians completing them at the bedside. Nurses reading or adding to them on the ward. Clinical coders translating the inpatient spell into the data the trust uses to get paid. Administrative staff using the same data to meet national clinical dataset reporting requirements.
One form, four very different sets of needs. Design it wrong and someone's job gets harder, or the trust loses money.
The trust also brought me on because of SNOMED CT, the clinical terminology standard used across the NHS. Getting the coding right isn't optional. It's how the money flows.
The work
Design the forms. Including the logic.
The priority was the inpatient clerking forms and discharge summaries, with everything else to follow. All of them existed on paper and all of them needed translating, which meant starting by walking through each form step by step with clinical stakeholders: how do you actually complete this? What works? What doesn't? What do you skip and why?
The whole time, there's a question running in the background that never quite goes away: if this field is too demanding, will it slow down clerking at the point of admission? And if it's too easy to skip, will something get missed, or worse, manipulated? That balance is the job.
Each form was designed in Miro, field by field: type, mandatory or optional, conditional logic, branching based on clinical findings. The kind of work that needs to be visual, because a clinician checking whether the logic reflects real clinical practice cannot do that from a spreadsheet.
Just over a year of work. Clinical clerking forms designed, logic documented, the trust already saving money on paper and printing costs, and data coming out of the digital forms clean enough to be useful.
Sound familiar?
Got something similar?
This kind of work is bespoke. It doesn't fit neatly into a standard engagement and it starts with a conversation, not a diagnostic. If you've got a similar problem, get in touch and we'll work out whether we're the right fit.
Get in touch