The debates around 7-day services, job contracts, scope of practice and emergency medicine overcrowding can sometimes feel like a maelstrom of fact, figures, spin and deceit. Even those of us working in emergency medicine find it difficult to determine the quality of what we hear through news channels and so it’s always good to read some science about really important issues that affect us.
A recent study in PLOS ONE has looked at the impact of extending primary care (GP) hours in Manchester, which also happens to be my home town. Over a one year period NHS Greater Manchester spent £3.1 Million pounds enabling groups of GPs to extend their services in the hope that they would reduce pressure on ED services.
This paper has already attracted interest and will no doubt be used by politicians and politicos to further their agendas. We on the other hand will be left to pick up the pieces.
They determined this by comparing attendance rates between GPs with the additional funding vs those without, and at first glance it looks like a success. The claimed rate of a 26.4% reduction in emergency department visits is a very tempting headline, but I was in Manchester in 2014 and I can assure you that we did not see fewer patients. The result is of course a relative risk and these are well known for being difficult to interpret and susceptible to making results appear more dramatic than they really are.
Let’s delve a little deeper from an emergency physician’s perspective.
- This is a really complex analysis and it’s tricky to find the raw data.
- They estimate that they had to provide 33,159 appointments to achieve 10,933 fewer ED attendances. So you have to offer 3 GP appointments to prevent a single ED visit.
- These were for minor problems (so it would not affect the more severe end of the ED)
- The analysis is complex as they are trying to compare reported attendance against trends but the suggestion of a 26.4% relative reduction is offset in my mind by the reported ED attendance rates per 1000 patients. In the control group it was 32.3 vs 29.4. Although statistically significant on their analysis I’m not convinced that we would notice.
- I am really struggling to understand how they came to the conclusion that there was a 3.1% reduction in ED attendances. It is stated in the paper but I just can’t see the baseline data to explain this (maybe it’s just me). They do state that it was a not a statistically significant finding.
- They do admit that a formal cost analysis was not done but do claim that costs have reduced in the EDs by £767,976 over the study period. I think this must refer to the ED costs WITHOUT taking into account the £3.1 Million spent. If I was reporting this I think I would subtract one from the other and call it roughly a £2.3 Million pound loss.
- No health outcomes were assessed and there was an assumption that admitted patients were ‘appropriate’ and not included in the analysis. I don’t like differentiating patients into appropriate and inappropriate on the basis of whether they were admitted. Anyone working in an ED knows that admission is not a good determinant of this.
- I would really like to see the data on ED attendances during the study period. Did this intervention really have any impact on what the local EDs saw in terms of patients numbers?
This paper has led to headlines suggesting that the funding really did cut ED visits such as this on the BMJ site, but I’m really not sure that it does.
What does this mean for us? It tells me that reducing ED attendance is complex and that simple measures such as extending opening hours do not always have the dramatic effect that politicians and some medical leaders predict.
It tells me that Emergency Departments provide really cost effective care, or does it. Rather it might just tell me that EDs are chronically underfunded and are providing care on the cheap.
I’ll leave it to you to decide, read the paper and get back to us.
Good luck with understanding the stats section.
BMJ Blogs Extending primary care hours cuts emergency department visits. http://www.bmj.com/content/354/bmj.i4818
Associations between extending access to primary care and emergency department visits. http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002113