NHS England’s instructions to hospital trusts on coding delayed discharges are ill thought out, unproductive and onerous. They completely ignore the findings of a major research and development project carried out in 2018-19. And they demonstrate that league tables and spreadsheets are not the solution to all the NHS’s problems. 

NHS England (NHSE) has instructed hospital trusts to collect data on ‘Reasons for Discharge Delay’. NHSE wants trusts to understand why it takes longer than expected to discharge some patients. But it has let itself be persuaded that what is needed is a list of ‘delay codes’, each indicating a single ‘reason’ for delay.

The designated hospital trust director or manager must identify patients who have suffered a ‘discharge delay’, choose from 37 delay codes the one they think best applies to each patient, and compile the figures into weekly situation reports (sitreps).

There is a fundamental flaw here. As an investigation by consultants Newton found in 2019: ‘Delayed discharges are the result of the way the system, as a whole, functions.’ THERE ARE NO SINGLE CAUSES OF DELAYED DISCHARGES!

NHS England is evidently oblivious to Newton’s findings, despite having joined in commissioning their work!

The coding scheme invents five categories of discharge code which are claimed to cover ‘the different stages in discharging a patient’ but does not say what those stages are! And it ignores the fact that decision-makers’ anxieties and beliefs can determine whether, for example, a frail elderly patient is moved to a care home. The well known aversion of some ward staff to allowing such patients back to their own home does not feature in any code.

So the thinking behind the scheme is badly flawed. And while compiling sitreps is an onerous task for senior staff in hospital trusts and local authorities – ‘Death by Excel’, as one director of local authority social services described it – there is at present no sign that NHS England is learning anything at all from them.

---

Contents

Introduction: the problem of ‘delayed discharges’

NHS England’s aims and methods: the ‘Excel approach’

Glossary (1): ‘Discharge process’, ‘Discharge delay’

Glossary (2): ‘Reason for discharge delay’

Glossary (3): ‘No Criteria to Reside’, ‘Discharge-ready date’

The five categories of ‘Reason for Discharge Delay’ codes

The 37 ‘Reason for Discharge Delay’ codes: an encyclopaedia of hold-ups

What the published data tell us – and what they don’t

Other ways to research delayed discharges (1): Talking to service providers

Other ways to research delayed discharges (2): Tracking patients’ care journeys

Conclusion: The message for Wes Streeting

* * *

Introduction: the problem of ‘delayed discharges’
For many years there has been concern about the number of beds occupied by patients who are experiencing ‘delayed discharges’ (formerly known as ‘delayed transfers of care’) in NHS hospitals.[1][2] The term is commonly used as a shorthand label to designate patients who have received treatment and been certified by clinicians as medically fit to leave the hospital, but have not left. They are sometimes described as patients ‘who ought not be there’.[3] Even though they themselves may wish for nothing more than to escape from hospital, they are evidently viewed by some as freeloaders.

This report examines NHS England’s most recent effort to get to grips with the issue: its attempts to identify and codify ‘reasons for discharge delay’.

NHS England’s aims and methods: Sitrep instructions (the ‘Excel approach’)
NHS England’s (NHSE’s) current approach to the subject of delayed discharges is to require each acute hospital trust to collect and submit various data to do with its performance. The latest version of its requirements can be found in Acute discharge situation report: technical specification,[4] issued in May 2024 to NHS acute hospital trusts in England. I refer to it here as the ‘Sitrep instructions’.

These instructions are intended to support completion of the Acute discharge situation report (sitrep). Annex C to the instructions tells acute hospital trusts that every week, their Sitrep compilers must assign to each patient who has stayed for seven or more nights and who has been identified as having ‘No Criteria to Reside‘ (NCtR), one of 37 specific ‘delay codes’. Each delay code denotes a single ‘Reason for Discharge Delay’ (RfDD). The sitrep, which has to be signed off by a director or senior manager, must show, on a Microsoft Excel Spreadsheet, the number of patients to which each code applies.

Experience of this painstaking approach has been vividly likened by one participant-observer, a director of local authority social services, to ‘Death by Excel’[5]: I refer to it here as the ‘Excel approach’.

Glossary (1): ‘Discharge process’, ‘Discharge delay’
The term ‘discharge process’ appears in Annex C to the Sitrep instructions, in a passing reference to the seven ‘A codes’ which ‘capture hospital process delays at either the beginning or end of the discharge process’.[6] Strangely, although the term clearly played a significant part in their thinking, the authors of the instructions do not say what they mean by it. Significantly, they nowhere instruct the reader how to tell where a patient’s discharge process begins.

The term ‘discharge delay’ too is used in Annex C to the Sitrep instructions:

It is important that system partners, both locally and nationally, have a good understanding of the reasons behind discharge delays in order to target action on reducing them. [7]

But again it is not explained what is meant by ‘discharge delay’. Nor is any distinction made between what in practice are different types of ‘discharge delay’:

1. Delay when nothing is happening to progress the patient (like waiting in a
queue for residential accommodation to become available).

2. Delay when time is being used constructively (as when arrangements are
in the course of being made to provide social care or community health
services on discharge).

3. Delay when time is being taken to resolve an issue (e.g. to to settle a
difference of opinion about a patient’s need for care after discharge).

A type 1 delay may affect patients very differently from a delay of type 2 or 3, especially if the patient is not being kept informed about their place in the queue.

Glossary (2): ‘Reason for Discharge delay’
The Sitrep instructions continue:

A reason for discharge delay (RfDD) should be selected [from the 37 alternatives on offer] when a patient with a length of stay of at least seven days remains in the acute setting after being assessed by their medical team as having no criteria to reside (NCtR). [8]

The Sitrep instructions also say:

For national reporting purposes, patients should only be counted as having one reason for delay. This should be the primary reason for delay and in the case of more complex discharges … should reflect a shared picture across all system partners involved, and partners should be involved where appropriate in advising on the correct code. This code should reflect the most recent reason for delay, which may change for a patient from week to week of submission. [9]

This instruction fails to acknowledge that ‘reasons’ are inherently subjective (different people may perceive different ‘reasons’ behind a situation) and should not be confused with ‘causes’, which are not. It also denies the reality that an event or situation will invariably be the product of more than one cause.

Glossary (3): ‘No Criteria to Reside’ and ‘Discharge-ready date’
What are the criteria to reside on which access to stay in that hospital bed depends? The Sitrep instructions say:

Having No Criteria to Reside (NCtR) is defined as when a patient meets none of the Criteria to Reside (CtR) [see the Table below] (unless a clinically warranted and justified exception exists) … The day on which a patient is assessed as having NCtR is also their discharge ready date (DRD).[10]

---

Criteria to reside[11]

Requiring ITU or HDU care? / Requiring oxygen therapy/NIV? /Requiring intravenous fluids? / NEWS2 greater than 3? (clinical judgement required in patients with AF and/or chronic respiratory disease) / Diminished level of consciousness where recovery realistic? / Acute functional impairment in excess of home/ community care provision? / Last hours of life? / Requiring intravenous medication > b.d. (including analgesia)? / Undergone lower limb surgery within 48 hours? / Undergone thorax-abdominal or pelvic surgery with 72 hours? / Within 24 hours of an invasive procedure? (with attendant risk of acute life- threatening deterioration).

---

This list of criteria was originally drawn up by the Academy of Medical Royal Colleges (AMRC) and published by HM Government and the NHS in August 2020, during the Covid-19 pandemic, in the second edition of the hospital discharge guidance.[12] To date it has remained unaltered.

As we see, every single one of the Criteria to Reside is to do with the patient’s clinical condition. None is to do with their physical or psychological well-being, or whether they will be safe and cared for at their destination. Nor is there scope for a clinician to judge the probability of a patient needing to be readmitted if their condition worsens, or the advisability of keeping them in for observation.

But the latest version of Hospital discharge and community support guidance also says:

Every person on every general ward should be reviewed on a twice daily ward round to determine [whether they meet the Criteria to Reside]. If the answer to each question (see the above table) is ‘no’, active consideration for discharge to a less acute setting must be made … [13]

‘Active consideration’ clearly leaves open the possibility that people other than clinicians, and facts and judgments other than the exclusively clinical, may influence the decision whether a patient is ready for discharge. So, for example, it allows for the non-availability of a suitable destination for a discharge-ready patient, on which the Criteria to Reside are silent, to be taken into account.

The five ‘parent codes’ of the ‘Reason for Discharge Delay’ codes
In the Sitrep instructions, the 37 ‘Reason for Discharge Delay’ codes are grouped into five categories termed ‘parent codes’ (A-E). They are listed as follows:[14]

‘A’ codes: ‘Hospital process’. These [seven] codes are said to ‘capture hospital process delays at either the beginning or end of the discharge process’.

‘B’ codes: ‘Wellbeing concerns’. ‘These [four] codes capture delays caused when wellbeing concerns or concerns about readiness for discharge are raised about a patient, by themselves or a third party, and the concerns are not yet resolved.’

‘C’ codes: ‘Care transfer hub process’. ‘These [three] codes capture delays that happen immediately after the patient has been referred to a care transfer hub (or its equivalent) and where action is required by the hub.

‘D’ codes: ‘Interface process’. ‘These [twelve] codes capture delays that occur after the care transfer hub has made a referral to the person(s) or organisation(s) who will arrange a patient’s post-discharge package. This could include NHS or community partners, the local authority, social care and/or housing colleagues. These delays are during the phase where the care transfer hub is brokering or discussing the patient’s discharge’, for example.

‘E’ codes: ‘Capacity’. ‘These [eleven] codes capture delays where the patient’s discharge destination, package or care is determined but the discharge is delayed due to capacity constraints.’ For example, ‘the required service doesn’t exist in the local system; it exists but has no capacity to receive/support the patient; or it has been sourced but is not immediately available’.

These five parent codes are a varied collection. Three are said to do with ‘process’, but – judging by their names – the other two, ‘wellbeing concerns’ and ‘capacity’, are not. ‘Wellbeing concerns’ and ‘capacity constraints’ are not located in time and and do not convey the sense of movement or development implied by ‘stages’ and ‘process’. Moreover, while the five parent codes may loosely be said to ‘capture’ delays, in the sense of ‘record’, what they clearly do not do is ‘cover the different stages in discharging a patient’, as the Sitrep instructions claim.

We can, however, make sense of the five parent codes if we think of them as ‘hold-ups to discharge’: all of them comfortably meet this description.

The 37 ‘Reason for Discharge Delay’ codes: an encyclopaedia of hold-ups
The 37 individual ‘Reason for Discharge Delay’ codes, listed according to the five categories A-E, are shown in the boxes below.They are reproduced here from Table 5 of the September 2024 Sitrep reports.[15]

---

A codes – ‘Hospital process’
Awaiting therapy review of need for supported discharge (A1) | Awaiting medical review of need for supported discharge (A2) | Awaiting referral to care transfer hub for supported discharge (A3) | Awaiting patient transport services (A4) | Awaiting medicines to take home, discharge letter or other discharge documentation (A5) | Remaining in hospital due to infection prevention and control restrictions (A6) | Awaiting formal decision to discharge (inc diagnostic test results) (A7).

---

The seven A codes are said to ‘capture hospital process delays at either the beginning or end of the discharge process’.[16] But ‘capturing delays’ does not help us to discover what caused them.

However, each of the seven codes does denote that something is holding up the patient’s discharge. In every case, an action by hospital staff is awaited. To lump together those seven actions as coming at ‘either the beginning or end of the discharge process’ is to imply a common feature that they simply do not possess.

The designers of the coding system are surely on the right track by looking to ‘process’ as an important key to understanding discharge delays. But their analysis is missing any concept of the patient’s care journey being a process of some kind.

At this point, however, we can note that all the A codes signify a delay during which nothing is actively being done for the patient. They are merely ‘awaiting’, possibly occupying a place on a waiting list, possibly in a queue. Possibly they are recovering after treatment and clinicians feel they need to stay in the ward, where their recovery can be monitored: such information is not collected. (The Sitrep instructions barely mention ‘treatment’, or subsequent recovery, reablement and rehabilitation, although treatment is the hospital’s prime function and a recognisable stage which every patient passes through.)

If hospital managers are to learn about ‘reasons’ for delay and to improve their hospital’s management system by addressing those ‘reasons’, it would help them to have information about which staff are involved. The Sitrep instructions tell us:

‘In general, a patient likely to have complex discharge needs is referred to a care transfer hub by ward staff, who begin discharge planning from the point of admission and present relevant information about the patient’s needs.[17]

But the A codes do not cover those activities of the ward staff. They are not designated as part of any process. Although some ward staff are habitually risk averse, and inclined to see older and frail patients as requiring institutional care, whatever their preferences and capabilities,[18] no A code will register this.

---

B Codes – Wellbeing concerns
Patient/family/carer concerns over discharge readiness (B1) | Ongoing safeguarding concern (B2) | Awaiting determination of mental capacity (B3) | Issues with discharge destination readiness (B4).

---

Despite what the Sitrep instructions say, the B codes are clearly not associated with a particular ‘stage in discharging a patient’. Indeed, wellbeing concerns may be voiced as soon as the patient is admitted, although they may be more likely to be voiced during their post-treatment recovery, as their eventual discharge gets closer.

As with the A codes, the patient’s departure is held up, although they are kept waiting not for an anticipated action, like those covered by the A codes, but in a state of uncertainty, waiting for the concerns to be addressed somehow, and outstanding issues resolved. None of the B codes signifies that steps are actually being taken to address concerns for the patient’s wellbeing.

Furthermore, it may well be that more than one concern is raised at a time, so the pressure to log ‘only one reason for delay’ may distort the perception or understanding of the patient’s situation.

---

C Codes – Care transfer hub process
Waiting for confirmation of immediate care needs and pathway (C1) | Awaiting necessary referrals by care transfer hub (C2) | Awaiting confirmation of funding eligibility (C3).

---

The C codes are characterized by the part of the organization to which the patient has been referred, the care transfer hub, and accordingly by the point in the patient’s ‘discharge process’ at which that hub is engaged.

Like the A codes, the C codes tell us that the patient’s progress is being held up, they are waiting for someone to take action. But they say nothing about actions that the staff of the care transfer hub may be taking. The C2 code suggests that they may be actively engaged in processing ‘necessary referrals’, but a C1 or C3 code suggests that they, like the patient, are merely waiting for others to act. Taken together, the three C codes offer minimal help to understand the causes of discharge delays and to enable action to reduce or eliminate those delays.

---

D Codes – Interface process
Home based rehabilitation, reablement or recovery service arrangements still underway (D1) | Other home-based social care service arrangements still underway (D2) | Other home-based community health service arrangements still underway (D3) | Housing provision arrangement for homelessness still underway (D4) | Bed-based rehabilitation, reablement or recovery service arrangements still underway (D5) | Residential/nursing home care arrangements still underway (D6) | End of life care inc Fast-Track CHC arrangement still underway  (D7) | Further action requested by agreed provider (D8) | Homeless with no recourse to public funds  (D9) | Self-funded care package arrangements still underway (D10) | Patient/family/carer choice discussions on package still underway (D11) | Out of area discharge arrangements requested, not completed (D12).

---

There are so many D codes not because the so-called ‘interface process’ is being addressed in greater depth than the ‘hospital process’ and the ‘care transfer hub process’, but because of the very wide variety of circumstances to be coped with and other organizations interacted with.

As we see, care transfer hub staff are engaged in both the ‘care transfer hub process’ and the ‘interface process’. In the former, as noted above, they are already interacting with people outside the hub (codes C1 and C3). So to treat the two processes as self-contained and entirely distinct from one another, as the Sitrep instructions portray them, is plainly unreal.

Finally, as with codes A-C, we can see that the ‘interface process’ is one during which patients are kept waiting, possibly in total ignorance, until brokering, discussing, arrangement-making, etc., with organizations outside the hospital, have reached a conclusion.

---

E codes – Capacity
Home-based rehabilitation, reablement or recovery services not yet available (E1) | Other home-based social care service not yet available (E2) | Other home-based community health services not yet available (E3) | Housing provision not yet available (E4) | Bed-based rehabilitation, reablement or recovery services not yet available (E5) | Mental health admitted patient care not yet available (E6) | Residential/nursing home care not yet available (E7) | End of life care inc Fast-Track CHC not yet available (E8) | Housing adaptations not yet completed (E9) | Capacity – Equipment and associated training not yet delivered (E10) | Awaiting restart of existing social care arrangements (E11).

---

The E codes all refer to a service by organizations, mostly outside the NHS, that is not yet available. These ‘capacity constraints’ are said to take effect towards the conclusion of the discharge process. But, like the B codes’ ‘wellbeing concerns’, capacity constraints cannot sensibly be said to constitute a stage in that process.

Like the B codes, the E codes are defined by reference to matters beyond the acute hospital that are claimed to be holding up the discharge of a patient.

Like the A codes and the B codes, none of the E codes signifies that action was actually being taken on the behalf of patients. The E codes signify only more ‘awaiting’ for the patient: time passing until the service becomes available.

* * *

It is noteworthy that more than four-fifths of the 37 delay codes describe the delay as ‘waiting’, including for actions ‘still underway’ or services ‘not yet available’ or issues ‘not yet resolved’. Of course, ‘waiting’ can never be a primary reason, or even a root cause. The coding scheme does not require the compiler to ask why the actions were still under way, or the services were not yet available, or the issues were not yet resolved. It is an extremely blunt instrument for discovering the root causes of a delay to a patient’s discharge.

What the published data tell us – and what they don’t
The raw data produced from the Sitreps are published monthly by NHS England as datasets.[15] Totals are provided for England, the seven English Regions, the 42 Integrated Care Boards and the 109 trusts. For each one, a figure is supplied for each of the 37 codes. The figures are presented each month in a single table with 159 rows and 37 columns: 5883 entries in all.

Given the limitations of the coding scheme and the rationale behind it; given too that we know nothing about the predilections of the individual compilers of the various Sitreps and any pressures that they may be under, and that no commentary has been published alongside the tables presented, it is not possible to draw any reasoned conclusions from the raw figures.

However, in an article in the Health Service Journal,[19] ‘The trusts struggling most with delayed discharge ‘interface’ problems’, Matt Discombe reviewed the first three months’ worth of data, for June-August 2024. His article highlighted ‘the wide range of reasons behind discharge delays’. He pointed out that at more than 20 trusts the majority of delayed discharges were caused by ‘interface issues’, while ‘almost a third [were] down to NHS and partners “brokering or discussing” patient care’, with an additional 30 per cent [being] due to hospital processes or care transfer hubs’. ‘At many other trusts, however, these “interface issues” made up a very small proportion of their delays – with issues relating to system-wide capacity, internal hospital processes and care transfer hubs being the bigger factors.’

But nothing that has been published so far shows that lessons have been learned. HSJ approached the trusts with the five highest proportions for ‘interface issues’, care transfer hubs and internal hospital process for comment. Two of them said they were working with system partners to reduce the delays. One added that delays have ‘dramatically decreased’ since the beginning of August, when it implemented its new care transfer hub. Another said patients are added to its care transfer hub ‘at an early stage in their pathway’, and therefore ‘we do not think it is helpful to refer to these data as representing ‘delays’’. And yet another said it did not recognise the data as a new IT system has impacted its ability to accurately report the numbers.

Tellingly, none of the trusts approached told HSJ what, if anything, they had learned from the exercise.

Other ways to research delayed discharges (1): Talking to service providers
The Excel! approach is by no means the only approach to researching delayed discharges that has been tried.

In the winter of 2022-23, Alex Baylis and his colleagues at the King’s Fund investigated the use of one-off funding provided by central government to reduce discharge delay in the winter months. They interviewed commissioners and service providers in six local areas of England to find out how they experienced the process of receiving additional funds, making plans, and delivering those plans.[20]

As part of their work they discussed with their respondents the ‘main reported causes’ of delayed discharge. These are listed in the box below.

---

Main reported causes of delayed discharge, from Alex Baylis et al[21]

Lack of staff | Poor hospital administration (eg, delays in discharge letters) | Delays with hospital drugs/pharmacy | Poor discharge planning | Non-compliance with discharge protocols | Risk aversion in hospital staff | Lack of hospital physiotherapy and occupational therapy assessments | Availability of transport | Lack of intermediate care beds | Lack of home support (eg, meals on wheels) | Lack of specialised services for people with complex needs | Poor care home handover | Delays accessing home adaptations or equipment The need to discharge to a wide range of local authorities with different processes | Homelessness | Lack of responsibility for self-funders | Patient choice

---

As we see, this list is very different from the list of 37 ‘Reason for Discharge Delay’ codes listed in the Sitrep instructions. Unlike the codes, they go beyond the ‘awaiting’ formula and employ the language of the respondents themselves. Hospital administration, discharge planning and care home handover are described as ‘poor’. And there was reported to be a ‘lack’ of staff, hospital physiotherapy and occupational therapy assessments, intermediate care beds, home support, specialized services for people with complex needs, and responsibility for self-funders. Poor administration and lack of staff are not among the reasons for discharge delay that can be cited in a sitrep.

So interviewing decision-makers can give a picture of the causes of delayed discharges that is very different from the picture one gets from Excel spreadsheets that use codes supplied by NHS England. It is a fuller and more revealing picture. Without being restricted to the prescribed language of the Discharge Delay codes, interviews can provide frank insights into the day-to-day management and operations of acute hospitals, and their interaction with local authority social services departments and the voluntary and community sector.

Importantly, skilled interviewers will learn about the influence of the local physical and social context of the hospital and local authority and the nature of the population that they serve. They will also learn about the cultures of the organizations they are studying. Baylis et al provide some revealing insights:[22]

Overall, we did not find that there was a common, shared understanding between all partners of the cause of delayed discharges in their system, nor a common, clearly understood set of priorities for tackling them.

Though we were typically told that relationships between partners were good, they were not always sufficient to develop a depth of shared understanding. In some cases, there were evident tensions between partners related to failures to turn these good surface relationships into meaningful action to reduce delayed discharges.

Cultural issues causing delayed discharge were also raised – in particular, a concern that hospital staff were ‘risk averse’ and over-inclined to believe that discharge to a residential or nursing home was required for some older patients.

Other ways to research delayed discharges (2): Tracking patients’ care journeys
In 2018 the consultant firm Newton was commissioned to carry out a project to examine the journeys taken by people through 14 health and social care systems across England. Their project covered a massive 10,400 patients. NHS England was actually one of the commissioning bodies. But when NHSE launched its delayed discharges coding scheme, its leaders appear to have been completely unaware of the work that Newton had done and reported on.

Newton’s report[23] outlines the methodology that they developed to identify ‘blocks and delays’ in these journeys. Working in partnership with local improvement teams across 14 local systems across England, the Newton team followed a four-stage approach: Engagement, Diagnosis, Implementation setup, and Design of implementation, transition and plan for sustainability.

The team looked at the journeys taken by people occupying 10,400 acute medical and surgical adult hospital beds. They asked how many were experiencing delayed discharges, and were able to identify blocks and delays in their journeys within each part of the system. This enabled them to determine what the causative factors for delays were.

The Newton team found that no single factor was ‘to blame’ for the delays. They asked a straightforward question: ‘What are people waiting for?’ and identified a ‘top ten’ list of ‘causative factors’ across all the areas that they studied. These are shown in the box below.

---

‘What are people waiting for?’ Top ten causative factors behind delays, from Newton[24]

Package of care | Social care assessment | Long term placement | Interim bed | Patient / family | Therapy assessment | Referral | Ongoing rehabilitation | Multi-Disciplinary-Team decision | Internal ward transfer

---

The question asked here could of course be worded as ‘What is holding up the patient’s discharge?’

The Newton team identified and analysed the outcomes for people whose discharge had been delayed. They concluded that there is no simple, single cause of delayed discharges. Factors contributing to delays are multiple, complex and vary significantly from one local system to another. NHS England’s delayed discharge coding scheme completely ignores this reality.

The Newton team assessed the effectiveness of cross-system leadership and governance for each system. Collaborative teams, which comprised Newton operational improvement specialists alongside local frontline clinical and social care professionals, discussed and designed approaches to reducing delays. Implementation plans were then developed jointly with staff from all partner organisations in each system: this was seen to be a way of gaining commitment to implementing the plans.

The Newton team found that, overall, some 2,800 of the 10,400 patients studied had been declared to be ‘medically fit for discharge’ (the criterion which preceded ‘No Criteria to Remain’) but remained in hospital. Case reviews conducted with practitioners revealed that of those 2,800 patients, between 32% and 54% were found to have been subsequently discharged to a setting where the levels of care were not well-matched to their needs. In around nine-tenths of those cases, the setting was providing a more intense level of care than would have been needed to maximize the individual’s independence. 

The Newton team reported that the patients who did best after discharge from hospital were those whose discharges were delayed the least. Those who did less well, having been placed in settings where levels of care were not well-matched to their needs, were found to have been delayed, on average, around twice as long as those who had done best.

The team came to the conclusion that to reduce the numbers of people delayed in acute hospital beds what was needed was ‘a robust, rigorous and systematic approach to measuring and tracking patterns of activity in order to identify the key issues contributing to delays, specific to the system’.

As noted earlier, Annex C to the Sitrep instructions stressed

It is important that system partners, both locally and nationally, have a good understanding of the reasons behind discharge delays in order to target action on reducing them.

NHS England’s delayed discharge coding scheme is incapable of contributing to any approach that involves tracking patients’ care journeys.

What we can see from the Newton report is that a great deal of understanding – together with skills and confidence – had been gained by the Newton team and the hospital trust and local authority staff with whom they worked. The survey of 10,400 patients clearly took significant funding: this was a major investment. But the management of NHS England appear not to have made the designers of the delayed discharge coding scheme aware of Newton’s work. In effect, they wrote off this investment and chose to start afresh.

* * *

At the opposite end of the scale to the Newton project was a pioneering study of just one patient’s care journey. It was carried out at Warwick Hospital by the South Warwickshire NHS Foundation Trust.[25]

A complaint from a patient called Gerry prompted the Trust to invite him to come in and talk to them about his experience. Gerry spent a total of eight days with them. He observed that, whilst his experience of the clinicians was excellent, there was no coordinated approach to his care, and communication between specialties was poor. The Trust mapped Gerry’s journey through the hospital and discovered that, during his 8-day stay, there were only 34 hours’ of value-added time. The rest of the time was spent waiting – for decisions to be taken, for tests to be carried out, for results to come through, for lists of To Take Out Medications to be compiled.

Based on Gerry’s experience, the Trust was able to improve their procedures, which they say enabled them to achieve significant reductions in mortality and numbers of medical outliers, and a rise in staff satisfaction levels. Their study illustrates some of the learning that a hospital trust can gain by examining individual care journeys, and the value of applying joined-up thinking and studying the care journey as a whole: a coding exercise based on the Sitrep instructions would have produced far less material from which to learn. (Incidentally, none of the South Warwickshire Trust’s figures in the HSJ Sitrep table appear outstandingly high or low.)

The Warwick study underlines the part that waiting plays in patient’s care journeys. But, presumably because Gerry was a mentally robust individual, it makes no reference to the psychological harm that long periods of waiting, with no control over one’s own life or knowledge of what is happening, can do to a patient.

Unlike the Excel approach and the coding scheme on which it is based, tracking or mapping of patients’ care journeys provides a means of gaining insights into factors affecting a patient’s care. It will also enable the application of Critical Path Analysis, a technique which hospital trusts could make more use of than they currently do.

Conclusion: The message for Wes Streeting
Wes Streeting, the Labour Government’s Health and Social Care Secretary, has made it known that he wants the NHS to achieve a substantial shift of its record-keeping and communications from analogue to digital, and to introduce ‘league tables’ for hospital trusts on the basis of measures of their performance. The experience of the Excel approach has warnings for both of these aims.

We don’t know how NHS England came to seize on the Excel approach, but it may be that the idea of employing digital coding of ‘reasons for discharge delay’ appealed to members of the NHSE Board as offering a quick and easy way of achieving an analogue-to-digital transition. Unfortunately one effect was that the Board offloaded to digital specialists their responsibilities for bringing the NHS into the 21st century. This study and the Newton report have shown that there were pitfalls that needed to be looked out for and avoided.

For the Excel approach to have been effective, it needed to have been designed by people with some familiarity with health care and social care organizations and experience of studying organizational processes: the designers of the Excel approach do not appear to have had this background. Even so, they should have read the Newton report before taking on their task.

As for league tables, we can say that while the discharge delay coding approach does superficially allow trusts to be placed in a crude form of ‘league table’,[19] so many qualifications need to be attached to it that nothing significant can be learned.

Despite the Sitrep instructions’ insistence on learning from the exercise, the Excel approach, as it was designed, ignored significant potential evidence in order to code what remained. It has proved to be an extremely blunt learning instrument in a situation where surgical precision was called for. The role of digital experts in updating the NHS’s data should not be left to them to figure out. It demands a high level of insight and expertise, including familiarity with the idea of ‘process’.

It appears that the responsibility assigned to the designers of the Excel approach ended with ensuring that tables of figures were produced. What was needed was not so much a ‘good understanding of reasons’ but a grasp among those who lead the NHS of the complexities of the ‘discharge process’ and the phenomenon of discharge hold-ups. This study suggests that such a grasp is as yet some way off.

Notes and references (All links to websites accessed 12 January 2025)

1. The King’s Fund, Delayed transfers of care: a quick guide, 4 January 2018
   https://www.kingsfund.org.uk/insight-and-analysis/articles/delayed-transfers-care-quick-guide
2. Peter Levin, Delayed transfers of care: Problems of definition, measurement and governance,
   3 August 2020
   https://spr4cornwall.net/delayed-transfers-of-care-problems-of-definition-measurement-and-governance/
3. Peter Levin, Overstaying patients in Cornwall’s main acute hospital: a problem that calls for
   careful monitoring and ongoing collaboration, 13 February 2024
   https://spr4cornwall.net/overstaying-patients-in-cornwalls-main-acute-hospital-a-problem-that-calls-for-careful-monitoring-and-ongoing-collaboration/
4. NHS England, Acute discharge situation report: technical specification, 31 May 2024,
   https://spr4cornwall.net/wp-content/uploads/NHS-England-Acute-discharge-situation-report_-technical-specification.pdf
5. ‘The reporting was onerous, it was death by Excel.’ Cited by Alex Baylis, Simon Bottery, Luca
   Tiratelli, Safya Benniche and Lillie Wenzel in Hospital Discharge Funds: Experiences in Winter
   2022-2023, the King’s Fund, 28 November 2023, p.33
   https://assets.kingsfund.org.uk/f/256914/x/ea8416a699/hospital_discharge_funds_2023.pdf
6. As 4, pp.9,12
7. As 4, p.8
8. As 4, p.8
9. As 4, p.11
10. As 4, p.22
11. As 4, p.22
12. HM Government & NHS England, Hospital Discharge Service: Policy and Operating Model,
    21 August 2020
    https://spr4cornwall.net/wp-content/uploads/Hospital-Discharge-Service-21-August-2020-Hospital_Discharge_Policy.pdf
13. DHSC & NHS England, Hospital discharge and community support guidance,
          26 January 2024, Annex D
    https://spr4cornwall.net/wp-content/uploads/Hospital-discharge-and-community-support-guidance-GOV.UK_.pdf
14. As 4, pp.19-20
15. These data are available from NHS England, which publishes them in a regularly updated
    Acute discharge situation report
    https://www.england.nhs.uk/statistics/statistical-work-areas/discharge-delays/acute-discharge-situation-report/
16. As 4, p.9
17. As 4, p.3
18. Local Government Association, Efficiency opportunities through health and social care
    integration, June 2016, pp. 4, 11, 15, 25-28, 35 https://www.local.gov.uk/sites/default/files/documents/Efficiency%20opportunities%20through%20health%20and%20social%20care%20integration.pdf
19. Matt Discombe, ‘The trusts struggling most with delayed discharge ‘interface’ problems’,
          Health Service Journal, 2 Oct 2024
    https://www.hsj.co.uk/quality-and-performance/revealed-the-trusts-struggling-most-with-delayed-discharge-interface-problems/7037906.article (£)
20. As 5
21. As 5, p.16
22. As 5, p.17
23. Newton, People first, manage what matters, undated (but 2019)
    https://reducingdtoc.com/People-first-manage-what-matters.pdf
24. As 23, p.22
25. South Warwickshire’s whole system approach transforms emergency care
    https://spr4cornwall.net/wp-content/uploads/South-Warwickshires-whole-system-approach-transforms-emergency-care.pdf