4 September 2024
Funnel plots in Excel
I want to present the data in a 'non-confrontational' way. I want to say to the consultants: "Here is the extent of the variation. And here are the consultants whose mean length of stay is statistically significantly higher or lower than the overall average." But I don't want to do it in a 'finger-pointing blame-game' way.
Read more...
2 September 2024
We've done data.
In my opinion, you're not 'doing' data about patient flow unless you are plonking the data right under the noses of the people whose behaviour needs to change in order for the improvement to happen. In fact, not just 'plonking it under their noses', but talking them through the numbers, engaging in a dialogue with them, making appointments to speak to them again after you've re-analysed the numbers according to their observations and comments.
Read more...
26 August 2024
The RCEM's Roadmap to Recovery
Acute hospitals need to make available to each specialty its indicators. Its actual numbers, its 'ought-to-be' numbers. The specialties need to have routine meetings at which these numbers are presented and discussed.
Read more...
23 August 2024
Meetings, bloody meetings
Suddenly, just before six o' clock yesterday teatime, I had a 'light bulb' moment: I've been looking at it the wrong way all this time. Flow_ology isn't a series of graphs; it's a series of meetings.
Read more...
13 August 2024
Using Data to Make Business Decisions
This is a course I've been delivering for a few years now. It was originally commissioned by the London Skills Development Network. Anyway, I've recently re-designed it to make it much more practical and example-based. It's now a hands-on course with quite a bit of Excel and PowerPoint whereas before it was much more theoretical.
Read more...
9 August 2024
Arguing with Numbers
I've re-written the Arguing with Numbers webpage to better reflect how the course now looks. It's developed quite a lot in the last couple of years as I've experimented with new ways of presenting data and encountered new books and videos on how to do it.
Read more...
6 August 2024
"It's data, Jim. But not as we know it."
For a number of years during the Pre-Pandemic Era, I used to travel up and down the country facilitating a workshop called Flowopoly. Flowopoly was a very physical, very tactile, 'anti-digital' way of representing a hospital's patient flow using tables, boards and cards.
Read more...
3 August 2024
Flowopoly gets a makeover
Flowopoly has had a makeover. It still begins with a whole-hospital replay, but it now comes equipped with 'flow moments', 'zoomed-in' downstream specialty replays (which allow us to see a fortnight's patient moves in 45 minutes), and simulated what-if scenarios, so that we can see the impact of changes we make to the system.
Read more...
2 August 2024
Dances, masks and feathers
A couple of days ago I was introduced to a meeting as a 'field analyst'. Which—rightly or wrongly—I took as a compliment. And which prompted me to re-visit and edit a piece about data analyst 'fieldwork' I wrote seven years ago.
Read more...
31 July 2024
Data League Tables
I had a conversation yesterday with a consultant geriatrician. One of the things we talked about was the pros and cons of presenting length of stay data in league table format.
Read more...
18 July 2024
The green zone
We can usually measure how long patients spend waiting in A&E for a bed to become available. The idea here is that we're just measuring the amount of delay. It's how long the patient is spending in A&E once the A&E bit of their treatment has been completed. This information can be useful if we want to get a clearer picture of the extent to which exit block is a problem for patient flow.
Read more...
18 May 2024
Red herrings
Was it Colonel Mustard with the lead piping in the Study? Or was it Miss Scarlett with the rope in the Conservatory? I was in Hatfield yesterday, teaching my Arguing with Numbers course to a group of people from Hertfordshire and West Essex Integrated Care System. One of the questions we grappled with was this: when you're using data to find out an underlying cause of something, and one of your investigations leads to a dead end, should you include or exclude that dead end when you write your final report?
Read more...
28 April 2024
The beds/data mismatch
Hospitals have specialties. And specialties have beds, which we tend to call 'bed allocations' or 'bed complements'. We gather data on the admissions, stays and discharges that take place in these specialty beds. But when we want to look at the inpatient activity in individual specialties, the bed allocations and the data often don't match up. It's important to address this mismatch. I believe that when we report inpatient data to inpatient specialties, we need to make clear the relationship between - on the one hand - the activity and - on the other - the space that activity is supposed to fit into to.
Read more...
27 February 2023
When to act on data
There's a diagram Nick Tordoff and I use in our Data Conversations course, and which I've now started using in my Using Data to Make Business Decisions course. It shows what it might feel like to be a decision-maker who has to make a decision (or take an action) based on data, and it's a representation of how they'd decide whether or not to trust the data, and how they'd evaluate the risks and benefits associated with any action they might take.
Read more...
23 December 2022
Operational versus strategic
It often feels as if we only ever address patient flow at an operational level. Our focus is on the hospital front door. Our attention is devoted to moving individual patients more quickly. Our actions are intended to speed up—rather than change—existing processes. But patient flow can only really be fixed if it is addressed at the strategic level. Less emphasis on what we do about Patient A and more emphasis on how we change the processes in place for dealing with all of the patients like Patient A.
Read more...
4 November 2022
Flavours of fullness
Whenever we use data to describe the fullness of a ward (or a specialty or a hospital), we usually reach for an average bed occupancy figure. The trouble is, this average figure is hugely deficient. Bed fullness is too multi-faceted a phenomenon to describe using just one number. Let's see if we can find a better way to capture all the dimensions of bed fullness.
Read more...