Going Beyond Analysis: Do We Need More Hospital Beds?

I found myself at the Midlands Analyst Network Huddle yesterday. The question on the table: Do we need more hospital beds? I was probably the least “analysty” person in the room. There I was, surrounded by people who can make spreadsheets sing, wondering what on earth I was doing there.

But again, I tend to gravitate towards wherever the mess is and if there’s a room full of people trying to make sense of a messy problem, I’m usually quite happy to pull up a chair.

I found the discussion fascinating. Across the NHS, hospital beds are under enormous pressure. Since COVID, that pressure has felt much greater than demand alone would suggest. The analysts explained why. For more than 30 years we’ve been doing more with less. Bed numbers have steadily fallen while admissions kept rising. We’ve managed it largely by helping people leave hospital sooner. In the 1990s, patients stayed for well over a week on average. Today, it’s often just a few days.

But you can only squeeze so much out of the system.

The people still occupying beds tend to be older, frailer and living with multiple conditions. Every extra day you remove becomes harder than the last. That’s good analysis, but, for me, it’s where the interesting bit begins. Because numbers are brilliant at telling us what is happening, they’re less good at explaining what it means.

Why does an approach that transforms one hospital barely make a dent in another? Why does every improvement seem to arrive carrying a new problem in its suitcase? Reduce admissions and demand pops up somewhere else. Shorten hospital stays and complexity increases. Run beds closer to full capacity and suddenly staff, emergency departments and winter resilience are carrying the strain.

It’s a bit like squeezing one end of a balloon. The air doesn’t disappear. It simply bulges somewhere else.

That’s why I’ve become increasingly convinced that analysing the mess isn’t enough. We also have to appreciate it. Because healthcare isn’t a puzzle waiting for the right person to solve it once and for all.

It’s a living system: the problems evolve, the context changes, people’s needs shift and what worked brilliantly last year may struggle this year. The goal isn’t to find the answer. It’s to find an answer that’s good enough, for now, in this context, while knowing we’ll probably have to rethink it later.

Oddly, I find that quite reassuring. It means complexity isn’t evidence that we’re failing.

It’s simply the price of working with people.

A few things I left thinking about

• There’s always another problem to fix. Health care, and human systems more broadly, don’t run short of things that need solving. That’s not a sign we’re failing, it’s just the nature of the work.
• Knowledge is partial, provisional and contingent. We never have the whole picture, what we know is always liable to be revised, and it depends on the circumstances we’re in at the time. Improvement, then, can only be defined against the local context. What counts as better in one hospital may not hold in the next.
• Expect temporary, contested solutions, not final ones. Because knowledge keeps shifting and context keeps changing, nothing we arrive at is likely to be the last word. The only thing that’s for sure is that nothing is for sure.
• Appreciating and naming the mess is real work. It’s not a delay before the real work starts, it is the real work. Formulating the problem properly, seeing clearly what’s actually going on, is often what makes any solution possible in the first place.
• Analysts need room for that as much as they need the data. The numbers tell you what’s happening. Sitting with the mess is what tells you what to do about it.