“You can’t change what you can’t measure”. This quote is often attributed to Edward Deming and more recently to Peter Drucker. There are various versions of this statement and there is a strong argument that you need to be very careful about what you measure[1]. For example, the NHS four week wait has in the past led to unwanted outcomes with staff focusing on those patients that can be seen quickest rather than the patients that have the greatest need[2]. So be careful what you wish for.
You don’t have to look very far to find evidence of that sentiment when considering NHS funding and the debilitating effects of excessive demand and limited NHS staff.
I think what we can agree on is that we need the NHS to be as efficient as possible because there is, and I suppose always will be, greater demand for health services than supply. So how do we help the NHS? One way is to support the NHS with a variety of measures that allow them to make informed decisions and improve their service.
Let's explore how this applies to a procedure usually referred to as percutaeous coronory intervention.
The combination of coronary angioplasty with stenting is usually referred to as percutaneous coronary intervention (PCI)[3]. According to the NHS website:
“As with all types of surgery, coronary angioplasty carries a risk of complications. However, the risk of serious problems is small. Complications can occur during or after an angioplasty. It's common to have bleeding or bruising under the skin where the catheter was inserted.
More serious complications are less common but can include:
damage to the artery where the sheath was inserted.
allergic reaction to the contrast agent used during the procedure.
damage to an artery in the heart.
excessive bleeding requiring a blood transfusion.
heart attack, stroke or death”[3].
The introduction of new technology such as SeQuent® Please NEO and reducing major adverse cardiac events (MACE) means it would be well worthwhile understanding the impact of these events at hospital level.
For example, we recently examined PCI secondary care activity in detail. The following charts show some extracts from our PCI report.
Chart 1 – Summary outcomes information for PCI procedures in NHS England over the last 4 years
Chart 1 shows the outcomes following PCI at a national level over four years ending in 20/21. The data can be broken down by hospital and in this case we chose to separate elective and non-elective care. For example, we could compare one specific trust with the national average. Let’s select Barts Health NHS Trust.
In Chart 2 below, we can see a more detailed view of the percentage of procedures that experience bleeds each year. The light blue bar represents the national average, and the dark blue bar represents Barts performance. Of course, it is important to think about these figures in the context. Barts is a tertiary centre for cardiology and therefore most likely, has more complex cases.
Chart 2 – Percentage of procedures that experience bleeds by year, Barts Health NHS Trust compared to the national average
One way to explore this would be to look at the number of diagnoses associated with patients undergoing care at Barts. Another way would be to look at the cost profile of patients.
In Chart 3 that follows, we show the cost profile for all London hospitals. These have been presented in two ways. The gauge chart, at the top, shows two pieces of data. The thin line you can see on the arch represents the national average. The green bar on the arch shows the performance of the organisation being scrutinised, as does the percentage underneath the arch.
Chart 3 – Cost profiles for London hospitals
The second way of looking at the profile of cost data is the stacked column chart where each colour represents a different cost band, as indicated in the key below the title. Focusing on Barts, which is the fourth hospital from the right, we can see that on average it deals with a larger proportion of patients at the higher cost end and below average numbers of patients where the costs are low.
By selecting Barts from the stacked bar chart we can further drill down to see that patients that cost up to £2,000 nationally represent 23.37% of the PCI population whereas Barts has a figure of 13.5%, as indicated in chart 4, below.
Chart 4 – Cost profile detail for Barts compared to national average for procedures up to £2,000 and £2,000 - £3,000
In chart 5 we make the same comparison for the higher end cost profiles; patients over £5,000 represent 6.4% as a national figure whereas at Barts these represent 15.4%.
Chart 5 - Cost profile detail for Barts compared to national average for procedures from £3,000 to £5,000 and over £5,000
When considering these results in relation to reducing MACE, with some work, we can now start to express the benefits in a way that relate directly to the needs of the patient population and the organisations we are working with.
We know this is a very powerful way of engaging customers in discussions around the patient pathway and improving it.
We have developed an interactive dashboard containing all this data on PCI and more. Get in touch if you would like to see it.
At SSB Research we love a bit showing off and we would be delighted to share.
[1] https://medium.com/centre-for-public-impact/what-gets-measured-gets-managed-its-wrong-and-drucker-never-said-it-fe95886d3df6 [2] https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1467-9299.2006.00600.x
[3] https://www.nhs.uk/conditions/coronary-angioplasty/