Published: November 2017
Authors: Jo Andrews, Ben Richardson, Hannah Farrar

Managing Delayed Transfers of Care (DToC) has been part of the fabric of NHS life for decades, but it is time to re-think our approach. DToC is a fraction of the patients who are delayed in hospital and could be cared for elsewhere, benefitting the system and reducing the harm that comes from prolonged admission.

In March 2017, NHS England published Next Steps on the Five Year Forward View. In the section on funding and efficiency, it states: “At present around 2500 hospital beds are occupied by patients who are fit to leave hospital but are awaiting social care, and an equivalent number is occupied due to delays in community health services”. We estimate the true number of people medically fit to leave as at least five times this.

Over the past two years, we have conducted bed audits across the Sustainability and Transformation Partnership (STP) areas we have worked with, building on the excellent work undertaken by Public Health in Devon. In total, these audits cover 8% of the acute bed stock, and 9% of the community bed stock nationally; we have also audited mental health beds in three STP areas. We have found 30% of acute beds and 36% of community beds are occupied by patients who are medically fit to leave. These patients, who need no ongoing medical attention, stay on average seven days longer in hospital, (and in some cases weeks or months) for potentially avoidable reasons. Extrapolating this to the national bed base suggests 28,000 beds could be freed if patients moved to a more appropriate setting of care no more than one day after becoming medically fit

DToC - Fig 1.png
Fig. 1: Beds audited by sector

The number of patients classed as Delayed Transfer of Care by comparison was 7,100, one-quarter of the number who are medically fit. Continuing to focus on the needs of only a subset of the patients who could be cared for in an alternative setting reduces the opportunity to move the resource to where it is needed - delivering care outside the hospital.

DToC - Fig 2.png
Fig. 2: Extrapolated scale of opportunity

So why is there such a difference in the numbers?

The concept of a ‘Delayed Transfer of Care’ originates in the Community Care (Delayed Discharges) Act 2003, and is tightly defined. Patients must be both clinically fit to leave and have been declared fit to leave by a multidisciplinary team. Finally, it must be safe to discharge them.

The aim of DToC was to identify individuals who are the responsibility of social services, and the Act introduced a financial incentive for local authorities to meet their obligation to provide services needed for the individual's safe transfer to a more appropriate setting, in the form of a fine payable to the healthcare provider. Self-funding patients are counted as the responsibility of the health sector.

The fining process required the health provider to have given due notice of a patient’s expected date of discharge, and subsequent confirmation that the date remained valid (section 2 and section 5 notice); a minimum period of 2 days was allowed from notification to qualifying as a delayed transfer of care. The act was replaced by the Care Act in 2014, which made few changes but removed the obligation to fine, and most areas have suspended the fining process as running counter to collaborative working. The default attribution of responsibility for DToC remains to the health sector.

So, the DToC process means it is perfectly possible to have two patients with clinically identical needs, and for one to be declared as a DToC and the other not, depending on what forms have been completed and when. It places undue emphasis on who is responsible, which even in the absence of fines does not encourage working together to meet the needs of the patient. Finally, the performance management of DToC numbers can create an incentive not to identify patients as a DToC at all, and therefore reduce the numbers reported.

Why does it matter?

In our relentless focus on DToC, we tend to miss one vital fact: keeping patients in hospital who do not need to be there is not a neutral act.

There is an extensive evidence base that shows the impact of hospitalisation on physical function through loss of muscle mass, risk of falls, pressure damage, acquired infection, and for those with dementia or cognitive impairment in particular, confusion and delirium:

DToC - Fig 3.png
Fig. 3: Harm caused by hospitalisation

The likelihood of harm increases with length of stay, especially in the over 70’s. Across four STP footprints we found the over 70’s made up the 74% of those who remain in hospital after they have been declared as medically fit. The time that patients are delayed for is significant too, with nearly half of our audited patients being delayed by four days or more, as shown in fig. 4.

DToC - Fig 4.png
Fig. 4: Duration of delay

Further analysis of beds across two STP footprints is shown in Fig. 5. The width of each bar on the histogram represents the number of patients fit to leave in each time interval; the respective number of bed days is represented by the area of the bar and displayed above it. Patients audited are on the xaxis. These audits included acute, community and mental health beds and revealed a median length of stay of 4-7 days, as well as the full scale of delay experienced by some patients.

DToC - Fig 5.png
Fig. 5: The distribution of delays in discharge

The long-term impact of this additional time spent unnecessarily in hospital is more difficult to quantify, but the loss of physical function, confidence and independence increase short and long-term care needs, which isn’t good for the individual and places an additional burden on care services.

It is clear that from the patient’s perspective we should be minimising delays in discharge, irrespective of whether they are classified as a DToC or not.

What is the size of the prize?

The potential impact of addressing this issue, for the individuals who are currently experiencing delays in their discharge and are therefore at risk of harm, is significant. Addressing it also benefits hospitals that are struggling to run efficiently, and the system as a whole, which needs to ensure the best value possible is being secured for every pound spent on health and care.

Acute hospitals cannot function effectively at current levels of occupancy. We have looked exclusively at patients who are medically fit to leave, but many patients experience delays in care before this point, an issue the Emergency Care Improvement Programme’s ‘Red2Green’ approach seeks to address. Expecting teams to work more productively and to eliminate these delays when their patients are scattered across a hospital is unrealistic. The impact on elective care is also significant: lack of available beds means whole operating teams being underutilised and becoming frustrated, and patients waiting longer for surgery. Waiting lists are increasing - an issue which cannot be effectively addressed without tackling this waste of resources. Of course, the most visible point of impact is at the front door: maintaining flow in Emergency Departments relies on appropriate beds for patients to be admitted to, and for clinical teams to have the capacity to provide timely care.

We estimate the annual cost of providing 28,000 beds at £3bn, based on a weighted average reference cost across hospital care settings of £299 per bed day. Meeting the needs of patients who are already medically fit to leave in a different way could release capacity. While some of this is required to address the operational issues faced by most acute providers, some of the resource can and must be used to fund the transformation in care required outside the hospital setting.

What’s the answer?

Our continued focus on DToC rather than the broader group of patients whose care needs could be met in another setting makes answering that question accurately very difficult. We know for example that delays due to ‘social care’ make up around one-third of the DToC numbers, but our evidence suggests this is an underestimate because it only captures those who are entitled to state funding. The true number is closer to one in two, encompassing patients waiting for assessment, placement and packages of care, and data published by the BMA indicates the numbers awaiting placement or packages of care is increasing year on year. Other reasons for delay include transfer to other NHS settings and internal processes.

The current position is that many services providing care and support in peoples’ own homes are fragile and unable to cope with current levels of demand, but the approach of managing demand by delaying access to these is, by keeping patients in hospital for longer than necessary, compounding the problem.

There is a clear case for moving resource from acute to community services, but the services required will vary. Working with clinical groups we have identified a range of interventions that address the barriers that prevent patients from leaving hospital in a timely manner. Solutions need to address the barriers to timely discharge in the hospital, at the interface with out of hospital services, and in the community, ensuring that care in every setting is safe and appropriate.

It is important to recognise this is not, and must not become an exercise in “shifting care” from one setting to another. New, targeted services are required, as well as the expansion of some existing ones, to meet the needs of patients who are currently unable to leave hospital. These needs must be met without the risk of harm that accompanies continued hospitalisation.

In each area, a deeper understanding of the needs of all patients who are struggling to leave hospital is required to inform new service models, and not just of those classified as DToC. Failing to recognise the needs of all patients means we will get the type and scale of services needed wrong, potentially missing opportunities to build new, innovative models of care.

Making this happen

We have taken a prudent approach in estimating the cost of current provision, and the figure will be even greater in those organisations where extra capacity is provided by large numbers of agency staff. In many cases, the opportunity is for Trusts to consolidate care so that these agency costs can be eliminated. This results in a mix of cash savings along with the increased capacity to help deal with activity pressures and waiting lists, especially given the impact current occupancy levels are having on elective efficiency in many Trusts. Developing the detailed local plan for the scale of services required, and what shift in resource is possible, are the first steps.

The reinvestment required in community-based services varies between different elements, and between areas depending on existing provision and the specific needs of patient groups. Our analysis indicates this is likely to range from 19% to 33% of the cost of hospital care, particularly where community-based services can be delivered at scale.

Discharging patients who are medically fit swiftly requires other changes, including clarity about who is responsible for the patient, payment mechanisms that support delivery of the care model, and data to support both the smooth transition of care between services and transparent reporting of performance. There is also an urgent need for collective effort to address the fragility of the domiciliary care market, and make working in that sector as attractive as similar work in the NHS, which does not appear to struggle recruiting to healthcare assistant posts.

More fundamentally though, systems to work together towards a common goal of ensuring people get the care they need in the most appropriate setting: the right care, in the right place at the right time. The welcome announcement in the March budget of investment in adult social care can start this process, but only if there is a better understanding of the services needed and a collective effort to deliver these.

We are spending scarce money on a pattern of care that harms patients. Without collective effort and commitment at system level to change this, recognising the widespread benefit of doing so, and driven by the clinical case that our current approach is causing harm, we will not make the changes in services, processes and thinking that are needed. Shouldn’t this be the top priority for every STP?


Read this report as a .pdf