Should we scrap the target of a maximum four hour wait in emergency departments?BMJ 2017; 359 doi: https://doi.org/10.1136/bmj.j4857 (Published 25 October 2017) Cite this as: BMJ 2017;359:j4857
All rapid responses
In response to Stephen Black's rapid response (ref 1) , I am surprised to be accused of misleading but again other readers are welcome to make their own conclusion reading my response (ref 2). However, he asserted that I had ignored "the much better Canadian study that shows a clear relationship between total time in ED and mortality"; I did no such thing. I have never disputed the observation that the longer the admitted patient spent in ED the higher the inpatient mortality rate for admission via emergency; I had however previously pointed out access block (as a main cause of ED delays and overcrowding) is more likely a reflection of a overcrowded hospital which is dangerous for any patient (ref 3).
Stephen Black has "worked as consultant to several health organisations on the performance of A&E departments and the 4hr target", but he may not know that while an occurrence is closely associated with an outcome, bringing in an intervention to reduce this occurrence does not mean the outcome is reduced.
The classic example is the CAST study (ref 4): patients who suffer from myocardial infarctions (MIs) have a high risk of sudden death (outcome), presumably due to arrhythmia (occurrence). Antiarrhythmics (intervention) successfully reduced the amount of PVCs, but consequently led to more arrhythmia-related deaths. Similarly arthroscopic knee surgery (intervention) specifically addressing degenerative knee arthritis and mensical tears (occurrence) does not resolve the knee pain (outcome) attributed to the arthritis (ref 5). And it is a fact that the 4-hour target had failed to demonstrate reduced mortality resulting from its implementation in the NHS.
Stephen Black and Drs Boyle and Higginson should also have realised that there is strong Australian research (ref 6, 7, as well as being referenced by Boyle and Higginson themselves) that achieving the 4-hour target is not required to improve inpatient mortality rate; compelling evidence suggests a range of 60-80% compliance to the 4 hour threshold can achieve maximum reduction of mortality rates, whereas the mortality rate sharply increase as compliance to 4 hour threshold head towards 90% and beyond (ie the 4-hour target).
Let me be clear on this: the 4-hour-target for the NHS is achieving 90-98% compliance of emergency department (ED) visits where door-to-departure (discharge or admission) occurred within the time-frame of less than 4 hours.
So far in my previous and current response, I have pointed out that:
- Excessive resources has been focussed on getting the patients out of ED within a time frame while inpatient beds and services are cut or frozen.
- Reduction in inpatient mortality can be obtained without the achieving the 4 hour target, by as little as 60% compliance or extending the 4 hour limit
- The quest to achieve the target has resulted in increased admission rates and creative managerial accounting to comply with the directive, hence contributing to the appearance of reduced mortality rates.
Inpatient mortality rates are (of course) the proportion of inpatient death out of all inpatient admission over a defined timeframe. Any 12 year old can tell you to reduce the rate, you either reduce the number of deaths or increase the admission numbers, the latter being a consistent observation of time-based target implementation across the globe.
These supporters of 4-hour target may claim we are talking about the same thing; I doubt so. While I am unable to change their steadfast defence of an arbituary target of 4 hours that has not resulted in reduced mortality rates in the NHS, I hope other BMJ readers will be open to ideas that may achieve similar results at less cost, by accepting alternative time targets (5 hours, for example, as suggested by Dr Campbell - ref 8) and improving hospital systems as a whole.
Certainly this chain of rapid responses has vindicated my use of the title "smoke and mirrors" to describe the tactics of some supporters of the 4 hour target (ref 9).
3. Goh S. Emergency department overcrowding and mortality after the introduction of the 4-hour rule in Western Australia. Med J Aust 2012; 197 (3): 148. doi: 10.5694/mja12.10828
6. Sullivan C, Staib A, Khanna S, et al. The national emergency access target (NEAT) and the 4-hour rule: time to review the target. Med J Aust 2016;359:354. doi:10.5694/mja15. 01177. pmid:27169971.
7. Staib A, Sullivan C, Griffin B, Bell A, Scott I. Report on the 4-h rule and national emergency access target (NEAT) in Australia: time to review. Aust Health Rev 2016;359:319-23. doi: 10.1071/AH15071 pmid:26433943.
Competing interests: No competing interests
I continue to be concerned by the assertions raised in Drs Boyle and Higginson's rapid response (ref 1).
Here they suggest "time based targets are associated with reduced mortality", curiously dropping the word "probably" from the original article (ref 2). When they originally wrote "improvements in performance against the standard were associated with absolute reductions in mortality in admitted patients", they omit the fact that the intervention in the Derby study (ref 3) was a 90% medicine bed occupancy target from June/July 2013; this is 2 years after the abolishment of 4-hour-target as a primary "standard" and "the only measure of care quality" (as it was in June 2010 ref 4), to be reinterpreted at 95% as a part of quality indicators dashboard. Boden et al has made no statement in their paper to justify the view "Performance against the four hour target was almost certainly a powerful lever in Derby to reduce medical bed occupancy and reducing mortality", particularly when the study authors had emphasised in their paper:
"After the introduction of relevant interventions to reduce bed occupancy in medicine, our principle findings are statistically significant differences in medical bed occupancy (level change), 4 h target performance and SHMI and crude mortality. Where this study differs from previously published research is that the association of bed occupancy on mortality has been studied after specific interventions were undertaken to reduce bed occupancy. This has not previously been studied and is potentially reproducible by other hospitals and Trusts."
Certainly the idea of improving bed occupancy had occurred before June 2013: in April 2013 the medical directorate of Derby Hospitals NHS Trust "aimed to create better flow within the acute medical bed base at a large acute NHS trust by increasing the volume of discharges, reducing occupancy levels, reducing length of stay and helping improve performance in the emergency department (ED)." (ref 5).
Boyle and Higginson had suggested that "a sensible manager confronted by lengthy waits in the emergency department would direct their efforts to reduce exit block so that emergency patients who need admission get a bed promptly", then most of the NHS managers and certainly most clinicians supporting 4-hour target should have identified bed occupancy the real target instead of focussing on getting patients out of the emergency department* and encouraging inpatient boarding. And they havent seemed to have done that for the last 13 years; mortality rates could have improved if they did.
[*Perhaps it is a Freudian slip for Dr Boyle and Higginson to use the term "exit block" rather than access block; afterall the care of an admitted patient is not defined by the patient 'exiting' out the ED into the ward.]
They continue to dodge the issue of increased emergency departments (EDs) admission rates as a result of the "time based targets", nor other explanations of why mortality rates may not have increased under these targets, including the possibility that increased proportions of low acuity admission diluting the effect of absolute mortality numbers. The idea that EDs nowadays have more tendency to admit patients when under pressure, particularly involving those who would have been monitored in ED without admission prior the 4 hour target implementation, is not unique to myself; the National Audit Office made similar assertions (ref 6).
Boyle and Higginson claimed I made "the elementary mistake in viewing the four hour target as purely an emergency department problem". Surely my entire rapid response (ref 7) and previous stance with Dr Boyle (involving inpatient boarding for example ref 8) can corroborate that that I have been pointing out repeatedly that despite the increased admission rates via EDs, reducing hospital inpatient beds and disproportionate resources thrown at EDs compared to the rest of the hospital structure, it is the entire hospital process that is helping to reduce access block via hospital overcrowding (not just ED alone), and hospital mortality rates.
Acute Beds in NHS England have shrunk by 25% from 132,000 to 102,000 between 2003/4 (implementation of 4 hour target) and 2016/17 (ref 9). Does this look like a rational move as part of the "whole system change" that Drs Boyle and Higginson wrote about? Would a sensible manager in the NHS dream up this measure supposedly as part of the measures to improve 4-hour target, knowing that it would unavoidably increase bed occupancy rates and access block?
Lastly they have alleged, "Dr Goh seems to be suggesting that we should tolerate raised mortality as long as admissions don't rise."
I made no such suggestion nor do I seek to do so. What I have been saying (and trying to get fellow readers to understand) is that the goal of 4-hours to admit or discharge as a benchmark does not necessarily improve care or mortality rates if anyone involved mistakenly views achieving the 4 hours time target as a benchmark. Worsening waiting time has not resulted in more deaths (ref 10) and this validates the removal of 4-hour target as a standard (as I have always asserted, for example ref 11).
3. Boden DG, Agarwal A, Hussain T, et al. Lowering levels of bed occupancy is associated with decreased inhospital mortality and improved performance on the 4-hour target in a UK District General Hospital. Emerg Med J 2016;359:85-90.pmid:26380995.
4. Lansley A, Secretary of State for Health, United Kingdom. [Letter to John Heyworth, President of the College of Emergency Medicine.] 2010; 21 Jun. http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/di...
Competing interests: No competing interests
At the beginning of this century, the then Prime Minister, Tony Blair, announced that health expenditure was to be increased. This was met with expressions of concern from HM Treasury. I have a clear recollection of Alan Milburn, then Secretary of State for Health, telling a television news reporter that the NHS would willingly meet Treasury demands that value for money would be demonstrated by the NHS meeting targets.
This is the background for the development of the A&E and other targets that were then set. Campbell argues and Boyle and Higginson confirm that emergency departments have been able to invest to ensure that the target can be met. Similarly, the surgical waiting time targets and the cancer referral targets have encouraged investment in those services.
There were no targets set for mental health services. So, even in 2005 when policies were known not to be being implemented a major health trust “was asked to make savings to make up for budget overspends by other trusts.” Under-provision of mental health services remains a problem to this day. That is how the meeting of targets could be funded.
The economist Charles Goodhart is eponymously associated with a Law that is summarised as “When a measure becomes a target, it ceases to be a good measure.” Waiting times were a good way to measure provision of healthcare but once targets were set this ceased to be the case.
1] Campbell P, Boyle A, Higginson. I Should we scrap the four hour target wait in A&E? BMJ 2017;359:j4857
2] Cole A. Mental health plans are not being fully implemented in England BMJ 2005;331:592
3] Dyer O. NHS trust plans to cut consultant posts to make savings. BMJ 2005;331:797
4] Strathern M. ‘Improving ratings’: audit in the British University system European Review 1997; 5:305-21
Competing interests: No competing interests
It is interesting that both Black and Guptha speak out in favour of employing Toyota thinking ("Would Toyota tolerate such inefficiency?" and "..everyone should be using a Toyota-like approach").
Toyota give full credit to Deming for their enormous success, and have followed his teaching more closely than almost any other organisation. In doing so, they have taken to heart his fourteen principles, the eleventh of which states, "Eliminate numerical quotas for the work force and numerical goals for management".(1)
Toyota have a reputation for efficiency and eliminating waste, as Guptha points out, yet have achieved this without the use of applying incentives to hit numerical goals (targets).
This seems like a paradox in today's "results-driven" world, and yet an understanding of why Deming said it and his alternative approach find echoes in many of the Responses. In his book, The New Economics, Deming points out that, "A numerical goal accomplishes nothing... What counts is the method - by what method?".(2)
In Santos-Peña's Response, methods include "color-coded triage" and "optimizing the available resources, especially human resources"; Guptha highlights the method to improve the speed of "completion of assessments for patients who are deemed to be stable for transfer of care"; Goh points out methods that include "better medical handover (especially interhospital transfer), better staff education and protocol-driven patient care, as well as transient increases in hospital staffing (or beds); Wadwha explains methods that involve "efficient processing"; and Black's suggested method is to "rethink the whole process of patient flow".
By all means, let's hope to achieve waiting times of less than 4 hours, for 100% of the patients (the original NHS target, by the way), but let's not force it by attaching monetary or career incentives. Treat such numerical targets as expectations, and do the hard work that makes sense to move in that direction. This will involve working through the relevant methods mentioned in these Responses (and by the authors Boyle and Higginson), monitoring our results to see how far we get and learning from this what should be the next method, or adjustment, required to improve further. Results should teach, not dictate. And managers should oversee this process as a condition of their normal wage, without the need for, or fear of, extra incentives. Perhaps then we will even approach 3 hours, 95% of the time, and make Toyota proud!
1. Deming W.E. Out of the crisis. Chapter 2. Massachusetts Institute of Technology, 1982.
2. Deming W.E. The new economics for industry, government, education. Chapter 2. Massachusetts Institute of Technology. 1993
Competing interests: No competing interests
The 4-hour target is controversial as the original head-to-head and the responses to it demonstrate. But the debate is misleading and often unhelpful as the participants ignore other known evidence and seek to justify binary positions rather than learning the useful lessons we should learn from the evidence we have.
Peter Campbell is basically right that the target causes distortions and gaming. But the question is whether that gaming overwhelms the benefits. He also argues that the target was not evidence based when it was introduced, but fails to ask whether scrapping the target could possibly be an evidence based approach either. He argues that the benefits could have been achieved without the target by implementing system improvements but why not do those alongside the target?
Shyan Goh attempts to debunk the proposition that faster treatment improves mortality, which is one of the best arguments for the target. He argues that the original argument is misleading but his own response is equally misleading. He thinks the Australasian studies are not good comparators, but ignores the much better Canadian study that shows a clear relationship between total time in ED and mortality; he argues that the recent UK study was about lowering bed occupancy not improving ED, but ignores the well-known fact that this is the single most significant way to improve the speed in ED; he complains that the evidence for the 4-hr target came after it was implemented, but, hold on, wasn't the evidence that doctors really ought to wash their hands between patients only available after some doctors started to wash their hands?
Goh and Campbell are also prone to make errors that are widely held by other commentators. For example, Campbell characterises possible gaming strategies such as stacking ambulances and admitting patients to avoid breaching the target. The first doesn't help game the target as the clock starts rapidly after the ambulance arrives whether the patient has left the vehicle or not. The second is unlikely to work as the subgroup of patients with the longest waits is those who are admitted (for several years the biggest cause of long waits has been the delay finding a bed and when this is true admitting a patient completely fails to game the target).
Campbell correctly recognises that the target is not something that the A&E department can achieve purely by its own actions and requires joined up thinking across the whole hospital. If patients are delayed due to a lack of free beds, the A&E is powerless to speed up the process. In this he recognises an insight which is frequently lacking in both commentary and in hospital management. But this poses another question: why is it that hospitals choose to game rather than improve? He is right that they do but wrong to assume that the target is the cause of this dysfunctional behaviour: a basic failure to adopt good operational management principles is the fault here. The best way to meet the target sustainably is rethink the whole process of patient flow (which needs to involve how hospital manage their beds and how community services manage the discharged patients who need community follow up). We have no evidence that encouraging better operational management in the absence of a target would be better than encouraging better operational management while keeping the target.
He is also right that everyone should be using a Toyota-like approach (as Deeming advised). Focussing on the target and nothing else is naive, unhelpful and Sisyphean. Everyone should be looking at broader metrics that tell us more about the underlying causes of long waits. I've been telling people that for years as have people like the ex A&E Tzar Matthew Cooke. We should routinely monitor the mean and median wait as well as the % under 4hr. We should look at the whole distribution of waits for different patient types (which is why I know that admitting patients to game the target won't work). We should be applying SPC tools to learn from changes and failures. Surprisingly, even believers in the target will be happy to agree with all the actions he suggests to drive improvement.
There is also more evidence that critics claim about whether the target works. There was little primary evidence about the likely clinical effects when the target was originally set, but there was plenty of evidence about how utterly pissed-off the public were about having to endure long waits in crowded A&E departments that didn't seem to care about waits at all. Longer waits lead to higher mortality (look for this study: http://www.bmj.com/content/342/bmj.d2983 not the Australian ones or the supposed UK analysis which never looked at mortality and was too small to spot the effect anyway). The introduction of the target clearly led to dramatically shorter waits in the UK and, contrary to some of the criticism, it was met nationally for several years (suggesting it wasn't grossly expensive or difficult and that it could be met again). We also have a case study in what happens when the target is relaxed or less emphasis is given to it: things get worse quickly. When the coalition government changed the standard from 98% in 4hr to 95% in 4hr actual performance fell quickly (to a level where fewer trusts were now meeting the "easier" target than had previously met the tighter one). The target may be a blunt instrument, but in a politically-driven culture it really does affect behaviour.
We need better analysis of the data and more attention to good operational management. NHS Improvement has recently introduced a useful tool for improving flow (see https://improvement.nhs.uk/resources/emergency-flow-improvement-tool/) but it doesn't provide the full set of historic analyses that would help to identify known bottlenecks (work to develop a broader such tool was underway with Matthew Cooke in 2010 commissioned by an SHA but was a casualty of the abolition of SHAs in the Lansley reforms). The leadership of the NHS has been slow to realise the importance of both good quality analysis and good operational management for actually driving improvement. What Campbell suggests as an alternative to the target is actually a good list of things that it is essential to do in order to meet the target.
In some important ways we are having the wrong debate. Both sides should agree that a more Toyota/Deeming-like approach to improvement is required to drive more sustainable improvement across the emergency care process (and don't forget, this is much broader than just what happens inside the A&E). We should be talking about how to achieve that not having an ideological argument about whether we should abolish the target.
Competing interests: No competing interests
We were guarded in in stating that time based targets are associated with reduced mortality, we were clear that these are associations, and are not causal. Dr Goh is right that the intervention in the Derby study was lowering medical bed occupancy. However, Dr Goh makes the elementary mistake in viewing the four hour target as purely an emergency department problem. A sensible manager confronted by lengthy waits in the emergency department would direct their efforts to reduce exit block so that emergency patients who need admission get a bed promptly. Performance against the four hour target was almost certainly a powerful lever in Derby to reduce medical bed occupancy and reducing mortality. Acheiving the four hour target really relies on whole system change and optimisation of inpatent discharges, a point Dr Goh makes.
Dr Goh seems to be suggesting that we should tolerate raised mortality as long as admissions don't rise. While there are lots of problems with the four hour target, we would encourage Dr Goh to consider it as a marker of whole system performance.
Competing interests: AB is Medical Director of Cambridge Clinical Management Analytics
Would Toyota tolerate an inefficiency that would see an equivalent of 3 wards lost to delay in assessments every month?
Dr Campbell in his response indicates that the target would lead to gaming as it will be perceived as a threat by staff. There are, however, numerous examples across the NHS where targets have been viewed as a positive driver in improving patient care. The idea that dedicated staff would indulge in gaming to meet any target is a bit perverse and does not reflect hard working NHS staff in a good light.
There are trusts who have between 600-700 days lost due to delays in completing assessments every month. If the average length of stay is 7 days then it would mean that an extra 100 patients could have been accommodated in that month which is roughly equivalent to 3 medical wards. Would Toyota tolerate such inefficiency? There would have been no reason for clinical staff to look up NHS statistics if not for the pressures on trusts to become more efficient which is a direct effect of targets. While it is possible that 95% would learn from the efficient 5%, it is very clear that there are numerous local factors in the NHS that play a significant role that cannot be accounted for in the learning that Dr Campbell proposes.
Competing interests: No competing interests
With the need to provide patient centered care, equitable access and cost containment, it is important to minimize operational separation of emergency departments (ED) from the rest of hospital. The patient’s journey is a continuum with all staff working as a cohesive and high functioning team with transfers between caring teams that is both efficient and comprehensive. With increasing burden on health care organisations throughout the world, there is a need to maintain targets not only in the time spent in ED but for other aspects of a patient’s journey in the health care system including length of stay.
Not everyone is a proponent of the four hour ED stay but by no means does the target define completion of a patient’s investigations or management plan or even an established diagnosis. It serves as a point in time of a patient’s course where, with effective systems, a reasonable provisional diagnosis, initial management plan and investigations and the need for inpatient versus outpatient care can be determined. There is evidence of better outcome with earlier transfers from ED to appropriate treating units (1).
Eastern Health has modeled its seven-day a week consultant driven general medical service for 4 years. As part of that process, full support to and integration with the ED has been integral so that patients are moved from the emergency department to the ward based teams efficiently. With continued senior clinician presence and decision-making, greater efficiency and sustained quality have been observed with an impression of better patient experience. Junior and senior medical staff have collaborated to maintain capacity to meet ever-increasing demand for ED services. There is explicit understanding that patients that transferred to the ward may at times become unstable (there are infrequent occasions of MET calls soon after arrival on the ward) and have incomplete investigations or management plans but it is rare that within that four hour interval a decision by a senior ED clinician will be overtly incorrect and necessitate change of admitting unit. Crucial to the success of this process is comprehensive clinical handover.
We feel that a majority of patients can be processed with a reasonable differential diagnosis and commencement of work up and treatment in 4 hours. Through efficient processing we are able to continuously create capacity and a safer environment for subsequent patients. We do not expect to achieve the 4-hour target for all patients and unstable patients and flow blocks in wards can be limitations. Meeting this target therefore at least 80% of the time may be more realistic and achievable. With high demand such as disease outbreaks, major trauma and disasters as well as staffing issues it is expected that this target may not be reached but these events are not incessant. Even so, there are opportunities for prediction, particularly with seasonal events that may allow health services to flex capacity, staffing and resources in advance to deal with these. It is time we reconsider that it is too hard to meet the four hour target but rather find more innovative, system based and collaborative approaches so that what we do achieves the best patient outcomes.
1 J. Singer, H. C. Thode, Jr., P. Viccellio et al., "The Association Between Length of Emergency Department Boarding and Mortality," Academic Emergency Medicine, Dec. 2011 18(12):1324–29.
Competing interests: No competing interests
Adrian Boyle and Ian Higginson in their defence of "the target of a maximum four hour wait in emergency departments" (EDs), suggested the following:
"On balance, time based targets are probably associated with reductions in mortality. Although a large UK database study found that implementation of the four hour target was not associated with reduced mortality, a single centre UK study showed that improvements in performance against the standard were associated with absolute reductions in mortality in admitted patients. Several Australian studies that have evaluated time based targets have shown reduced mortality associated with introducing a time based target. New Zealand’s six hour target has also been associated with decreased mortality. Certainly, it does not increase mortality, and, perhaps counterintuitively, it does not increase attendances at emergency departments."
I am concerned the casual reader may be misled by their assertions.
I wish to point out the single centre UK study (ref 1) had the intervention as lowering levels of bed occupancy (and NOT the introduction of 4 hour target); the study authors concluded "lowering medical bed occupancy is associated with reduced patient mortality and improved ability of the acute Trust to achieve the 95% 4-hour target."
I can hardly call that an endorsement of the 4-hour target, much less the allegation that "improvements in performance against the standard were associated with absolute reductions in mortality in admitted patients" as suggested by Boyle and Higginson.
As for the other Australasian studies (ref 2-5), they all come with some important characteristics Boyle and Higginson do not mention: the 4 hour target (called national emergency access target NEAT) in Australia, 6 hour target in New Zealand are both associated with increased admission rates, varying from 9 to 24%, after the introduction of the target in EDs.
There is a possibility of hospital staff gaming the system to achieve this target, particularly when funding is directly involved in meeting performance indicators (Ref 6). For example, there is evidence of increased proportion of short-day admission and reduced length of stay. There may be various explanations how this can happen in the presence of increased admission rates; there are better management and discharge planning via inpatient service. Alternatively more patients may be admitted into the hospital to meet the target when it appears their presenting problem cannot be sorted out and discharged within 4 hours; before the target existed, these patients would have been discharged directly from EDs, albeit this may take longer than 4 hours to organise home services and followup. This would have been a compelling alternative reason why the patients under the 4-hour-target have a shorter length of stay than would occur with an admission to an inpatient bed.
This may also explain why the absolute number of inpatient deaths did not increase significantly with 24% increase in admission (Ref 4). It may even be possible to attribute this as due to a significant proportion of the increase in admission number possibly being low-acuity and without the same risk of dying as a typical admission case pre-target.
Furthermore, alternative mechanisms (other than the 4-hour rule) may account for reduced access block and related mortality rates (Ref 7). In the Western Australian study, access block rates had been declining for 5 months before the 4-hour rule program was introduced, and for 12 months before “solution implementation”. There are evidence that improvement programs involving better medical handover (especially interhospital transfer), better staff education and protocol-driven patient care, as well as transient increases in hospital staffing (or beds) may have occurred around the same time as the implementation of NEAT.
I am not sure if Boyle and Higginson realised the irony when they wrote:
"The four hour target is a simple, well understood measure that drives flow throughout the whole urgent care system. However, suggesting that it is no longer relevant is like changing the rule because you don’t like the result. There is no realistic alternative."
Prior to the announcement of the 4-hour target (by 2004) by the Department of Health in 2000-2001, very little evidence exists whereby a time-based target of 4 hours would improve outcome. The decision to use 4 hours as a standard of care (and a KPI target) is almost certainly a political one; they could easily have used 3 hours, 6 hours or any number out of thin air on a whim.
Much of the 'evidence' for 4-hour target came AFTER the introduction, and at the end, the "implementation of the four hour target was not associated with reduced mortality" by their own admission. New Zealand has demonstrated that a similarly arbitrary target of 6 hour can achieve the same results (ref 5). One can hardly say "no alternative exists to keep emergency departments working"! System change within the inpatient care process and discharge management are the real reasons why hospitals coped with increased admission from the 4 hour target.
Thus the focus on "4-hours" is unreasonable, expensive and wasteful; it should be changed not just "because you don’t like the result" but because the real improvement is what happens throughout the hospital system, and not only at EDs.
Time to stop the illusions and delusions; time to be practical and sustainable in austerity.
1. Boden DG, Agarwal A, Hussain T, et al. Lowering levels of bed occupancy is associated with decreased inhospital mortality and improved performance on the 4-hour target in a UK District General Hospital. Emerg Med J 2016;359:85-90.pmid:26380995.
2. Sullivan C, Staib A, Khanna S, et al. The national emergency access target (NEAT) and the 4-hour rule: time to review the target. Med J Aust 2016;359:354. doi:10.5694/mja15. 01177. pmid:27169971.
3. Staib A, Sullivan C, Griffin B, Bell A, Scott I. Report on the 4-h rule and national emergency access target (NEAT) in Australia: time to review. Aust Health Rev 2016;359:319-23. doi: 10.1071/AH15071 pmid:26433943.
4. Geelhoed GC, de Klerk NH. Emergency department overcrowding, mortality and the 4-hour rule in Western Australia. Med J Aust 2012;359:122-6. doi:10.5694/mja11. 11159. pmid:22304606.
5. Jones P, Wells S, Harper A, et al. Impact of a national time target for ED length of stay on patient outcomes. N Z Med J 2017;359:15-34.pmid:28494475.
6. Goh S. Lessons from the 4-hour standard in England for Australia. Med J Aust 2011; 194 (11): 615.
7. Goh S. Emergency department overcrowding and mortality after the introduction of the 4-hour rule in Western Australia. Med J Aust 2012; 197 (3): 148. doi: 10.5694/mja12.10828
Competing interests: No competing interests
The NHS, being an interlinked and interdependent system of health care, absolutely requires joined-up thinking to improve waiting times of patients in Emergency Departments. Ensuring faster assessments for Transfer of Care should be part of this approach. Dr Guptha has highlighted how there is enormous variation between the hospitals, and interpretation and response to such variation is the starting point for a scientific approach to improvement which will involve, as described in the "Yes" to scrapping targets part of the debate, suspecting individual causes for the performance of the 5% outliers which is to be dealt with (or learnt from), followed by support to improve the entire system for the remaining 95%.
Impetus to focus on Transfer of Care Assessments only needs political and managerial willingness, availability of time to do so (a commodity I suspect is in extremely short supply), and good data to guide decision-making. Why should it require a 95% target that only threatens the staff involved and promotes more game-playing behaviour such as, I could imagine, the opening of a Pre-"Transfer of Care Assessment" Unit that takes more people out of the statistical data?
Competing interests: No competing interests