Desire paths: what might a landscaper see in the health system?

There’s a small rectangular field near my house that has gates in opposite corners. A path runs around the edge of the field, meaning that to get from one gate to the other, you have to walk along two sides of the oblong. Naturally, no one wants to do this; it’s a pointless detour. So there’s also a dusty track running diagonally across the field – the quickest way from A to B – that has been carved out by hundreds of people simply walking the same route.

These types of tracks are called ‘desire paths’, and at a conceptual level they create an interesting lens through which to examine systems. Desire paths are things that occur when the system, in this case the design of the paths, doesn’t match up with what people actually want. They are a kind of visual representation of the wisdom of the crowd versus the fuzzy thinking of the system architect.

If we take this lens then and apply it to the health system, what might we see? Where are there examples of patients and communities stomping around the pathways designed by management – and what can we learn from this behaviour?

The obvious example is A&E. By some estimations (although the exact figure is contested), as many as 40% of people who show up at A&E don’t ‘need’ to be there. Anecdotally, this also seems to be an issue that the general public is aware of – rather than just an assessment that hospital managers share behind closed doors. And yet… people keep on coming!

The health system tends to view this phenomenon as a problem that needs solving. And of course we do need to reduce pressures on emergency services. But what if we look at this ‘problem’ as a communication from the public, telling us something about their desires? When services such as 111 exist, what is it about that kind of experience that people find unsatisfying, and makes them willing to put up with a trip and a long wait in order to see someone in person? What kind of non-emergency care needs are people showing up with, and what does that tell us about other kinds of service provision? A growing number of people showing up with dental issues is certainly something that needs examining in this light.

Another area of the health system where I’ve been thinking about desire paths is primary care. In conversations with GPs over the past year or so, I’ve heard about two growing trends in the patients they’ve been seeing. One is the ever-increasing number of people who have self-diagnosed conditions that they believe they are living with (particularly around conditions such as neurodiversity) and are going to the doctor in order to make it ‘official’. The other is people asking for various kinds of general ‘health checks’, despite not having any particular symptoms that they’re concerned about.

There are problems with both of these things, and for the most part, GPs seem quite frustrated with their growing prevalence. However, once again, if we apply a ‘desire paths’ lens, I think what we see here is a broad want from people to take a more active role in the management of their health and to be able to see doctors not just to talk about sickness when things go wrong but to discuss their health in a more holistic, day-to-day kind of way.

Now it may well be that these are things that GPs, stretched as they are under present circumstances, are simply not able to give people. But the fact that that desire is being manifested is something that the health system as a whole should take seriously and think about how it can respond. If GPs are unable to give people what they want in this space, who can? Is there an opportunity to think about digital tech? It’s worth considering at this point how much online content and advice there is for things such as nutrition, exercise and wellbeing. The NHS attempts to provide this too, but it’s tiny compared to other actors here – is this something that the system should look to change?

When it comes to desire paths, as a system architect you basically have three choices. One is to ‘pave over’ the path and therefore make it an official part of the system design. Another is to block it, stopping people from going the way that makes sense to them and instead forcing them to use the system as you designed it. In some instances, there may well be reasons to block people. There may be hidden risks that the public can’t see if they go a certain route and therefore a necessity to protect people. But if this isn’t the case, then we need to think about what we can learn from people’s revealed preferences in systems, and about how we can listen and respond to the non-verbal messages that patients and communities are sending us. And that leads us either to ‘pave over’ or to our third option, which is to take the information that we learn from the desire path and go back to the drawing board to design something better.