Two cars with an average speed of 40 mph.
Imagine you are flying down the highway in the car of your dreams, sunglasses on, the wind in your hair, great music playing and an empty road straight out of a car advert. But then the engine makes an awful sound and the speedometer drifts down from 60mph to 50, 40, 30, 20…10. You pull over to the side of the road and pop open the bonnet, painfully aware that anything more than a screen-wash refill will probably require somebody who actually knows what they are doing.
Focussing on the speedometer. It’s a pretty crude marker of car function; it tells you nothing about tyre pressure, fuel efficiency or oil level. On the other hand, going fast is pretty much what cars do. A modern car that cannot achieve 60mph is probably broken.
The Current Standard
In the UK, a 4-hour standard for hospitals was introduced by the government in 2004. It’s been through a couple of iterations, but the current version in England states that at least 95% of patients attending Emergency Departments should be seen, assessed and either admitted or discharged within 4 hours of arrival. This applies to all patients who attend the department, no matter what is wrong with them, or how they get there.
The introduction of this standard was not universally welcomed, but it has revolutionised Emergency health care within the UK. Introduced alongside substantial new funding for hospitals, the service improved to the point that patients no longer spent hours or even days on hospital trolleys in corridors, waiting for beds in the hospital wards. The limited amount of research in this area suggests that reducing ED crowding has improved outcomes for patients including mortality.1,2
Stephen Powis, the NHS National Medical Director, was asked by the government to review the NHS Access Standards in 2018. He has documented his findings in this document that was published in December 2020.3 It is out to consultation until 12th February 2021.
The Situation in UK EDs
The authors set out the issues with ED performance as follows:
- ED attendances and hospital admissions rates have risen.
- The total number of hospital beds has decreased.
This mismatch between supply and demand has led to poor flow of patients, resulting in crowding in Emergency Departments. Crowding is known to be associated with increased morbidity, mortality and length of stay in hospital for affected patients.4,5
This paper sets out to change how the activity of emergency and urgent care services are measured. The intention is that by changing how the emergency services and urgent care services are perceived and accessed, services will be improved, and risk reduced.
Issues with the Current Standard
The document lists the following concerns with the current 4-hour standard:
- The target does not measure total waiting times
- It does not take account of the patient condition.
- Whole system performance is not measured
- Clinical advances in Same Day Emergency Care (SDEC) are not considered
- The target is not well understood by the public
- There is significant variation in the proportion of admitted patients across the country
To take each of these points in turn:
- Waiting times are already recorded but do not form part of this target.
- Patients are triaged on arrival so that the sickest patients get treated first but this is not target driven. Patients with some specific conditions e.g., stroke, major trauma, myocardial infarction or sepsis already have their times and outcomes monitored with other national clinical standards already in place.
- Arguably the four-hour standard does provide some measure of whole system performance as a system that is performing well it would be expected to achieve this standard. Failure to achieve the standard demonstrates issues with the system without identifying where they lie.
- Same Day Emergency Care (SDEC) is the new name for the older practice of managing patients as out-patients for conditions that would once have resulted in admission, e.g., cellulitis or suspected deep vein thrombosis. Patients managed in this way are included in the numbers for the performance standard but are not examined as a subgroup.
- Healthwatch UK, an independent group set up by the government to assess the views of healthcare users, performed a survey in 2019 which informs this statement.6 In a survey of 1,700 people, only 46% recognised that the four-hour ‘clock’ started at registration, though 90% of people did identify that the clock started at some point between arrival and initial triage. 70% said that an average waiting time would be easier to understand. There is some evidence that people tend to overestimate their understanding of common concepts (https://doi.apa.org/doiLanding?doi=10.1037%2Fxge0000161) and one wonders how deep their understanding is.
- Variation does exist, but it is unlikely that the four hour standard is entirely responsible.
Another concern with the four-hour standard is that people ‘game’ the target, as evidenced by the disproportionately large number of patients who are discharged or admitted in the golden ten-minutes between 3 hours 50 minutes and 4 hours. This is likely to apply to be true of any time-based target.
Most of the larger EDs have Observation Wards where patients, who are likely to be discharged, can wait for blood tests or other investigations to be completed, or for those who require a longer period of observation. For example, a patient with a swollen calf may be admitted there pending an ultrasound scan. Following the scan, the patient would be discharged with appropriate treatment and follow up based on the scan results.
Observation wards avoid crowding in the EDs, offer the patients a more comfortable environment and allow rapid discharge without becoming embroiled in the bureaucracy that can result from being admitted as a specialty in-patient. It also stops the clock, in terms of the four-hour standard.
Proposed Changes to Measurement
The proposed changes to measurement are spread over four ‘services’; Pre-hospital, Emergency Departments (A&E), Hospital and Whole System.
- Response time for ambulance
The time taken from calling an ambulance to the crew being with the patient, is a standard that is already recorded as a national quality indicator for ambulance services in England.
- Reducing avoidable Emergency Department attendances
This is also already recorded as an ambulance service quality indicator where crews assess a patient and find an alternative to hospital transfer. It may also be valuable to record the proportion of patients who call back or attend hospital at a later time.
- Proportion of contacts via NHS111 that receive clinical input
Presumably this will be of value to service providers though it is not clear from this document whether a high number or a low number is desirable.
A&E (Emergency Department)
- Percentage ambulance handovers within 15 minutes
This is time taken for the formal handover from the ambulance crew to the ED staff and is already recorded.
- Time to initial assessment
This is the time taken from the patient booking into the department to having some form of clinical assessment; this would usually be the triage time. This time is also recorded already.
- Average time in department for non-admitted patients.
It is interesting to base a target around a definition that can only be confirmed retrospectively; it is not always clear if patients will require admission when they first attend. Separating this group from patients who go on to be admitted to hospital implies independence from the issues that affect patient flow in Emergency Departments.
- Average time in department for patients who go on to be admitted to hospital
- Clinically ready to proceed
The time at which the patient has been assessed and treated and is ready for admission or discharge, previously known as the ‘decision to admit’ time.
- Patients in the department for more than 12 hours
The document states that there is no valid reason why patients should have to spend more than 12 hours in the Emergency Department although it’s not clear why this number was chosen rather than a shorter period such as 6 or 8 hours.
Standardising the point at which the clock starts, i.e. when the patient is booked into the department, would be a very useful step. This may appear obvious, but the current 12h standard has no defined start point. Some centres use the patient arrival time, while other hospitals only start counting after the patient has been seen by a specialty or when a bed is requested. This means that patients can spend more than 24 hours in a department without officially breaching ’12 hours’.
- Critical time standards
This refers to time-related clinical standards specific to patients with four conditions: stroke; acute myocardial infarction with ST elevation (STEMI); major trauma and RAPID (acute physiological deterioration).
For patients who present with these conditions, there are already national clinical standards in use and the relevant times are recorded. A concern with focussing attention on specific conditions is that this may be at the cost of neglecting other conditions of similar severity. For example, patients with non-ST elevation MIs have a similar in-hospital mortality as patients with STEMIs.
The Eleventh Standard: A New Offer to the Public
Within the document is a table providing a list of the clinical standards that are being offered to the patients, written in plain English. An eleventh clinical standard is included here although it does not appear to be mentioned elsewhere in the text: ‘Where your clinicians think you need to stay in hospital after your initial care, you will be moved to an appropriate bed within one hour.’ Arguably, achieving the patient flow necessary to meet this single standard would have the greatest effect in terms of patient care.
Critical Time Standards
The justification for the introduction of the Critical Time Standards is that ‘the highest priority patients will get high-quality care with specific time-to-treatments, with proven clinical benefit.’
However, the conditions chosen are all already subject to national clinical standards associated with specific times. One concern is that such targets could focus attention onto patients with these specific conditions at the cost of patients with equally serious, but unlisted conditions.
The text provides two examples of improvement in clinical outcomes, secondary to specialised pathways (clinical standards), though both are slightly disingenuous.
The authors state that there has been a 19% improvement in survival rate since the inception of the major trauma network in 2012/13. The reference is not provided in the document but is listed in the related document, ‘Clinically Led Review of the NHS Standards’.7 The paper referred to is by Moran & Lecky et al (2018). This paper actually found a 19% increase in the odds of survival, rather than the probability of survival. This sounds petty, but the chances of survival were 92% when the network was started, or odds of 11.5 to 1. Multiplying these odds by 1.19 increases your odds of survival to 13.7 to 1, which corresponds to a survival probability of 93%. To be clear, I think that the UK Major Trauma network has massively improved the care that this group of patients ; the mortality benefit is important but less dramatic than implied. Ed – we have previously questioned the validity of this increase in trauma mortality on a St Emlyn’s review here.
Another example of care improvement resulting from clinical standards is for patients with strokes where the authors state that the 30-day mortality for patients admitted with stroke has decreased from 27% in 1998 to 17% in 2010 to 13.6% in 2015/16. Again, the references for these figures are not provided in this document but are provided in the related document, ‘Clinically Led Review of the NHS Standards’.
The first figure is not based on UK numbers but refers to a relatively small study based in South London.9 The second and third figures are from the SSNAP Audit which looks at data submitted to the audit group from stroke centres in England, Wales and Northern Ireland.10 It’s difficult to obtain national figures now as the Office of National Statistics has changed the way it provides information about these patients but the Nuffield Trust has published data for these patient groups over this time period and states the following.11
“In the UK, 30-day mortality after admission to hospital for ischaemic stroke (based on linked data) decreased from 18.3 per 100 patients in 2008 to 11.6 per 100 patients in 2013, but the rate has since stayed roughly constant. In 2017, the UK had the second highest 30-day mortality rate of the comparator countries.”
“In the UK, the 30-day mortality rate after admission to hospital for haemorrhagic stroke (based on linked data) decreased from 37.1 per 100 patients in 2008 to 30 per 100 patients in 2013 and has remained roughly constant since then. The UK’s mortality rate is relatively high compared to other OECD countries.”
Presentation of New Clinical Standards
While the acceptable thresholds for each standard have not been disclosed at this point, the intention is to have a binary result for each of the ten standards. Each hospital would then score zero or one for each variable up to a maximum of ten, presented as a summary result.
Do the new standards work?
It’s very difficult to say as very little information has been forthcoming. Fourteen sites in England were chose to be pilot sites while a further nineteen Trusts were picked to be controls; the document states that the hospitals were selected to provide a good mix of hospitals but there does not appear to have been any form of randomisation or statistical justification for their selection.
Very little data was provided in terms of results. The number of patients waiting in departments for more than 12h from time of attendance did appear to reduce more in the test sites compared with the control sites, but the denominators are not provided.
|Baseline||May to Oct||Difference||Percentage reduction|
The conversion rates (admission rates) are also provided but only as percentages, so the significance of the changes shown cannot be assessed.
|Baseline||May to Oct||Difference|
The mean time in the departments have also been provided through three of the test sites were excluded. Again, very limited data has been provided: average times without range or median values, or numbers of patients.
|Baseline||May to Oct||Difference from baseline||Percentage change|
There are apparently some minor discrepancies in the table, but the only change of note appears to be the increased time that patients in the test sites wait to be discharged; the significance of this cannot be calculated from the provided data.
The data also shows that patients requiring admission tend to wait much longer than four hours, no matter which hospital they attend.
Given that so many of the times of interest are already recorded, it is disappointing that no analysis has been provided in this paper as to how changes to the standard would relate to ED performance ratings. Sarah Scobie has compared how the new clinical standards would look against current standards using data from the last few years in this article for the Nuffield Trust.12
Setting standards of care, and comparing performance against your peers, are two great drivers that can improve medical practice. Periodical assessment of these standards also works, but an evidence-based approach should be applied to changes in practice just as it would be applied to the introduction of a new treatment or diagnostic test
The four-hour standard may be an oversimplification of a complex system but it can be applied directly to an individual patient rather than a population. It is achievable, and simple enough to be encapsulated in a single sentence. The consequences of applying the new standards, and dismantling the old, are unclear.
The new standards are out to consultation until 12th February 2021 so please take the opportunity to express your opinion here.
- Sprivulis PC, Da Silva J-A, Jacobs IG, Frazer ARL, Jelinek GA. The association between hospital overcrowding and mortality among patients admitted via Western Australian Emergency Departments. Med J Aust [Internet]. 2006;184(5):208–12. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16515429
- Geelhoed GC, de Klerk NH. Emergency Department overcrowding, mortality and the 4-hour rule in Western Australia. Med J Aust. 2012;196(2):122–6
- Sprivulis PC, DaSilva JA, Jacobs IG, et al. The association between hospital overcrowding and mortality among patients admitted via Western Australia emergency departments. Med J Aust 2006;184:208–12.
- Guttmann A, Schull M, Vermeulen M, et al. Association between waiting times and short term mortality and hospital admission after departure from emergency department: population based cohort study from Ontario, Canada. BMJ 2011;342:d2983.
- Moran, Christopher G et al. “Changing the System – Major Trauma Patients and Their Outcomes in the NHS (England) 2008-17.” EClinicalMedicine vol. 2-3 13-21. 5 Aug. 2018, doi:10.1016/j.eclinm.2018.07.001
- Variations in Stroke Incidence and Survival in 3 Areas of Europe, Charles D.A. Wolfe, MD; Maurice Giroud, MD; Peter Kolominsky-Rabas, MD; Ruth Dundas, MSc; Martine Lemesle, MD; Peter Heuschmann, MD; Anthony Rudd, FRCP for the European Registries of Stroke (EROS) Collaboration.
- Alan Grayson, “Trauma in the UK, who cares? St Emlyn’s,” in St.Emlyn’s, December 31, 2018, https://www.stemlynsblog.org/trauma-in-the-uk-who-cares-st-emlyns/.