The COVID-19 crisis really has brought out the differences between risk takers and the risk averse, demonstrating which of us cautiously follow all the rules and which of us treat them as polite suggestions, to ignore as we please.  In this blog post I’ll try to rehabilitate the reputation of the risk takers and share my hope that the Digital Health Ecosystem Wales will encourage risk takers and the risk averse to work together in harmony.

Image of 5 white paper boats and 1 red paper boat

photo of huw angle, authorThe NHS as an organisation is risk averse, no question. You could probably consider that its USP, because who wants to take risks where public health is concerned?  The current crisis, however, is one of those moments where the biggest risk is to do nothing at all.  So, NHS organisations have been forced to take some calculated risks and do so quickly and decisively. 

For NHS Wales Informatics Service (NWIS), these calculated risks have been things like rolling out remote working, quickly integrating with new partners and rapidly widening the user base for certain clinical applications.  It’s been fascinating to watch as many projects are pushed on at a pace that would previously have felt downright reckless.

When the risk averse are forced into taking risks, one of two outcomes are likely.  The first is blind panic and certain doom, the second is a calm deep breath as they summon forth their inner maverick, that risk-taking hero they’ve been cultivating for situations just like this. But how do you cultivate such a hero in a “safety first” environment where projects must be agreed by a multitude of stakeholders before passing an untold number of reviews, form-filling exercises and rubber stamp procedures? Boundary pushing, that’s how.

Make sure you’ve always got mavericks pushing at the boundaries, starting with the boundaries of functionality. The reason is that anything new and unknown is inherently risky, and a maverick will embrace that risk with a pioneer spirit, learning more by pushing the boundaries further. Obviously, there must still be limits on risk, but we can make these “risk budgets” a whole lot more generous than we do for existing “safety first” functionality.

The second boundary we want mavericks pushing at is rules and red tape, the stuff that many projects may silently suffocate in. A maverick will scream for help, decrying the insanity of filling in 10 pages of “Form A” when it’s identical to “Form B”.   If things look dire, a maverick won’t be afraid to make some calculated scissor-cuts through that red tape, spinning up unofficial development teams and commandeering a server to make sure their new app is delivered in time for pilot.

I believe that much of this maverick behaviour happens already in the NHS, but informally, unsanctioned and unacknowledged.  Without it, I feel sure the NHS response to COVID-19 would have been much more like blind panic and certain doom. The problem with this boundary pushing being done informally, however, is that afterwards the gains are often lost, forgotten or worse - left neglected to become a future liability.

Many tech companies acknowledge and formalise this kind of boundary pushing with agile workflows.  Sprints and SRE style error budgets provide a safe space for mavericks to experiment with new functionality. Regular retrospectives, deployments and use of CI/CD pipelines help to keep the red tape well pruned. These are exactly the approaches we’ve aimed to adopt with our work on the Digital Health Ecosystem Wales programme.

Our aims for the Ecosystem are to lay the foundations for simpler integration with and within NHS Wales, starting with a Developer Portal and API platform.  Even just getting started with that work has required a fair amount of boundary pushing, redefining roles and team structures to better fit with a more agile development process, experimenting with new models of collaboration and communication between industry and the NHS and taking advantage of the cloud for developing and deploying applications. Some of these things are still something of a novelty for NWIS, and a real puzzle when it comes to formal assurance.

My hope is that over the next year, we’ll look back at these early frontiers as familiar territory and see mavericks at NHS Wales as key members of agile teams, pioneers who are well provisioned to explore and map out ways we can improve healthcare with technology.  As much as we may enjoy the safety of home, we never know when we’ll be forced to look beyond it, maybe leaving the map completely and venturing off into the unknown.


Some Examples of COVID-19 Boundary Pushing

I spoke to people working on projects in response to COVID-19 and asked them what boundaries they’ve seen pushed to accelerate their work.

Phill Cann – COVID-19 Mortality Document Sheet

Phill helped develop the e-Form used to record deaths from COVID-19, an important tool for providing daily numbers to Public Health Wales, so clearly, they needed a release as soon as possible.  Patient safety was still paramount, but Phill says that they were willing accept minor data integrity issues for the first roll out.  They treated it as an MVP, Minimum Viable Product, and then improved it with frequent releases afterwards, adding validation and swapping out freeform text fields for pre-defined dropdowns.

MVPs are not unheard of in NHS Wales, but the bar is often set so high that it might be an acronym for Maximum Viable Product. Phill’s example illustrates the worth in assessing the payoff of lower reliability targets against the potential pain. Data issues can be painful, but we can’t avoid 100% of them, what we can do is set targets and regularly review them against the cost of achieving them.