If you surf to a news website right now or flick on the TV news this evening, you might for a moment think that you watching a bad science fiction movie of a dystopian future starring your local A&E department. Sadly, it is real. The news has gone mad about A&E and the waiting time crisis that it faces. This morning the BBC reported that only 82% of patients are meeting the four-hour waiting time target. It doesn’t make pleasant reading.
It also isn’t easy to understand what is happening or why. Terms such as ‘inappropriate use of A&E’ and ‘trolley waits’ are thrown about without much explanation. I am going to try and explain some of the complexity around what is happening in A&E and why I think we are looking in the wrong place for improvement. Hopefully by the end of this blog you will agree with me. But don’t feel you must.
Inappropriate use of A&E
Just to be clear, I don’t particularly like this term, but there is a lot of talk about inappropriate use of A&E. Seven-day access to GPs has been promoted as a solution to inappropriate use. If we look at datasets such as the Hospital Episode Stats, we can see that between 30-40% of patients are discharged from A&E with no treatment or advice only. These are patients that could potentially be seen by an alternative health service such as GP surgery. As a caveat, this figure has been hotly disputed; although I do not know of any public data to support that argument. If you have some then please get into contact.
So, problem solved? Well not really. Out of those patients who inappropriately attend A&E we estimate that between 15 – 25% would be willing to see a GP on a Saturday or Sunday. That is, at best seven-day access has the potential to reduce A&E usage by 10%. When you do the arithmetic, and add in other potential users of seven-day GPs such as 111 referrals and the odd random walk-in appointment then our estimate falls to a 3% reduction on average. This estimate agrees with work our sister organisation GM-CLAHRC work looking at seven day access. In terms of the what public say that they want from GPs, we can look at the survey of general practice. This found that a ~80% of people do not feel weekend opening is an issue, but that some people such as the young or those that work might find it useful. Opening on a Sunday is unlikely to improve access.
The headline figures are not the full story
When we read about our local A&E we are often presented with a figure such as 80% of patients were seen within four-hours. These figures can be broken into two chunks. Many A&E departments separate patients into two groups each representing a level of emergency. Patients who walk into an A&E are often seen in a ‘minor’ emergency area. The processes here are relatively simple and minor patients have the best chance of being seen within 4 hours. Some patients can also be dealt with outside the A&E in what is called ambulatory emergency care. Minors and ambulatory systems can make efficient use of space and resource. But crucially they have relatively little reliance on inpatient beds that are a very scarce resource within a hospital.
The same cannot be said for major emergences. These are the people who are most likely to wait over four hours. You can see some of these poor individuals in the recent dystopian news stories covering the A&E crisis. This is a depressing picture of people lay on trolley beds in corridors or, even worse, outside in a large ambulance called a jumbulance. This is happening because the hospital process is gummed up. It is just like when you fill up your bathtub and leave the taps on. To avoid the bath overflowing you need to periodically remove the plug. The same is true for A&E. Typically, between 30 and 35% of major emergencies require an inpatient stay. There needs to be regular movement of these patients from A&E to in-patient beds or else the hospital overflows. And in this case, we are not talking water on your newly tiled floor.
Once we breakdown the results like this we can see that targeting patients with minor aliments is not an answer to the A&E crisis. We need to find a way to help patients in majors.
We have been looking in the wrong place
We need to acknowledge that the A&E crisis is a systemic issue. I recently went to a physio appointment where it turned out the pain in my arm was caused by a problem in my back. Massaging my arm helped a bit, but it just alleviated the symptoms for a short time. The pain was back with reinforcements the following week. The A&E crisis is similar; it is a symptom of wider problems.
At one level, we can consider how an A&E interacts with the rest of a hospital. We know that as hospitals progressively get more full there is an almost exponential increase in admission waiting times and A&E overcrowding. It’s clear then that reducing average occupancy is paramount. To put it another way this means that some of the beds need to be empty for some of the time. This reserve capacity helps absorb the peaks in demand. To free up capacity hospitals are focussed on patients who are in beds who no longer need to be there. We know that this group of patients has grown substantially over the last few years. This brings us to the second level of systemic thinking: what happens outside the hospital affects the number of beds required.
The NHS publishes information on people in beds who no longer need to be there in the delayed transfer of care statistics. The chart below plots the areas where we have seen the biggest growth in delays: the setup of care within patient’s own home and finding a nursing home place. To demonstrate that there hasn’t been growth across all areas I have also plotted patient and family choice. This has fluctuated but not seen growth like the other two areas. Before we jump to the conclusion that this is simply a capacity problem in social care and nursing homes, I want you to think about bath tubs again (sorry). There are two ways to control the level of water in a bath tub, as a bath has taps (inflow) and a plug hole (outflow). Reducing the water inflow by even a trickle can have a large cumulative effect over time. In the case of hospitals, there is always going to be a tipping point where an individual requires a visit to A&E and subsequent inpatient treatment, but we must not forget that it is not just about getting people discharged. I believe that many NHS and social care organisations recognise this, but often the most visible and tangible area to tackle are delayed transfers of care.
 |
| Fig 1: No. of patients with delayed transfers of care by month. Source. NHS England. |
This is not to say that the processes within an A&E could not be improved. Having spent time in A&E as part of a research project, I am convinced that some micro-optimisation can be done and would help alleviate the pressure. But, just like my bad arm, it is a temporary fix that will flare up again in a few weeks’ time. The singular focus is not helpful. We are looking in the wrong place for solutions.
So, what can be done?
Let’s be clear. No one has the answer at this point. So, the first step is to beware of media savvy talking heads selling snake oil. I do believe that we are now able to start to rule out some so-called solutions. Or at the very least to acknowledge that the benefits of some proposals do not include reduced A&E use. This isn’t the same as solving the A&E crisis, but it will allow us to move on and think through what else might help.
Throwing in my two cents, I would say we need to make better use of data and analytics (but then I would say that because that is what I do). I was intrigued to see that a recent report commissioned by NHS England recommended the creation of an NHS DataLab. What a brilliant and exciting idea. Classic improvement work focuses on introducing a new service to solve the problem. This is not sustainable and in my opinion makes the system more complex. A big focus of data science work to alleviate the A&E crisis could be on dis-investment in services and re-investment elsewhere; for example, in managing frailty and multi-morbidity as opposed to single diseases. This won’t be easy, but the reality is that there is no more money. I would also hope that any NHS data science initiative will not forget mental health, social care, nursing homes, voluntary services and self-directed care.
Sorry for the long blog. It is complicated issue. This is just the tip of the iceberg.
Dr Tom Monks leads the Methodological Hub here at NIHR CLAHRC Wessex –
The Hub provide methodological expertise to the CLAHRC programmes of research and responsive modelling support for NHS partner organisations. The Hub constitutes six core posts of differing methodological expertise including qualitative research, statistics, clinical informatics, operational research and health economics
clahrcwessexnews 