The impact of delayed discharge fees


Title: Survey on the Implementation of the Community Care (Delayed Discharges) Act (2003)

Authors: David McCoy, Allyson Pollock, Christina Bianchessi and Sylvia Godden

Institutions: University College London and the University of Edinburgh

Title: Can We Tell if Government Policies Are Working? Interpreting Government Data on Delayed Discharges from Hospitals.

Authors: Sylvia Godden, David McCoy and Allyson Pollock

Institutions: University College London and the University of Edinburgh


These two linked publications arise from research funded by the Economic and Social Research Council. Both focus on the impact of the Community Care (Delayed Discharges) Act 2003 which gave NHS hospitals the option to charge the then social services departments (SSDs) a daily tariff for patients who continue to occupy a hospital bed after being passed medically for discharge due to delays in the provision of social services. The first report describes a survey of English SSDs to discover how local parties had reacted to this power and with what consequences the second focuses on the validity of the official data that are used to measure delayed discharges.


For the first report a survey was conducted to examine the extent to which hospitals implemented the charging option, to estimate how much SSDs paid acute hospitals, to compare the SSD payments to hospitals with the delayed discharge grant, and to explore any associated changes in the ­working relationship between SSDs and hospitals. In the second report, the researchers summarise data on the situation before the legislation, describe the current data-collecting arrangements and assess their validity and usefulness.


For the first study, a questionnaire was sent to all SSDs in England in November 2005, and this resulted in a response rate of 53% (80 authorities) – more than adequate for this methodology. Of the 80 respondents, 49 reported not having made any reimbursement in 2004-5, and the total paid to hospitals by the remaining 31 councils amounted to £4.6m. Within the latter, payments varied from a few hundred pounds to several hundreds of thousands. The Department of Health had created a new grant for councils – the delayed discharge grant – to help them pay for reimbursements, but the hospital reimbursements amounted to less than 10% of the new grant. Indeed in 2004-5 only two councils paid out a sum greater than their delayed discharge grant – Brighton & Hove and Southend-on-Sea.

The arrangements in different localities were put into two main categories. First, those where the hospital is reimbursed and keeps the money as a source of income (about one-third of cases) second, an arrangement where the hospital does not earn income from delays in discharge (the rest). The latter implies an arrangement where there is no financial incentive for the hospital to charge SSDs as much as possible, and includes agreements that any charges paid are re-invested in community-based services. SSD respondents reported that working relationships with hospitals had improved in just over half of all cases, and had deteriorated in only 7.5% of cases.

The second study takes a long sweep on the accuracy of data on rates of delayed discharges, noting that reliable figures on volume and frequency have been hard to come by – various past estimates had ranged from as low as 4.8% to as high as 27%. However, in 1999 the Department of Health introduced situation reports (Sitreps) to collect this data.

With the 2003 act, Sitreps was modified in an attempt to standardise data collection and improve the accuracy of the figures. Using data between September 2003 and July 2005, the researchers found a decrease in delayed discharge bed days each quarter from 93,761 to 66,248. SSDs were estimated to be responsible for just over a quarter of cases of delay, most ­commonly due to waits for a residential placement and delays in providing home care packages.


Delayed discharge is back in the news again, with the successes reported in the ­immediate wake of the 2003 Act not being sustained. Figures released by the Department of Health in November seem to show a rise of 30% in the number of bed days lost from 2005-6 to 2006-7. This has led to the resurrection of long-standing arguments about the extent to which hospitals are discharging people much more quickly than in the past but not transferring resources to social care to support the rehabilitation process.

What can we learn from these two reports? The first lesson concerns the need for collaboration rather than conflict – not an easy task when the government puts so much emphasis upon competition. The finding from the first of these reports is that in most cases hospitals and SSDs chose to avoid a system of charging and payment, probably in the belief that both would benefit more from a co-operative approach. However, this was in the days before payment-by-results and before the kicking-in of the heavy private finance initiative repayments now facing many district general hospitals – the policy context is less favourable towards partnership approaches now that hospitals are under huge financial pressure to maximise income.

The second lesson concerns the validity of data on delayed discharges. Within a week of the Department of Health conceding that the number of bed days lost through delayed discharges in 2006-7 had broken through the million barrier, care services minister Ivan Lewis had changed his mind and said the real figure was 733,000. U-turns of this magnitude do nothing to inspire confidence in the validity of official data, and the second of these reports helps us to understand the limitations of Sitrep data:

There are differences in the way delays in discharge are defined as a consequence of variations in medical opinion about the readiness of a patient for discharge.

The organisational culture of hospitals and PCTs may influence where thresholds for safe and appropriate discharges are drawn.

There have been no attempts to standardise the Sitreps data even though hospitals vary in effort and ability to collect accurate and comprehensive data.

Some hospitals and SSDs may under-record the frequency and volume of delays in order to appear better organised and receive higher performance management rankings.

Sitreps does not collect person-centred data. Therefore, it is not possible to say whether totals are due to a few patients experiencing long delays, or many patients with short delays.

Underlying these specific limitations of Sitreps is the more fundamental point that a reduction in delays in discharge will not constitute evidence of overall success if it is accompanied by premature discharges resulting in readmissions, poor post-discharge care, a reduction in patient and family dissatisfaction, and a shift in the cost of providing post-acute health care to social care, patients and their families. Currently there is no way to officially monitor these issues accurately.

Bob Hudson is visiting professor of partnership studies at the School of Applied Social Sciences, University of Durham


Both of these reports form part of the Public Services Programme Discussion Paper Series which aims to disseminate innovative and, as yet, unpublished writing in order to generate public service centred discussion – in other words it offers an early glimpse of unfolding work.

The first of the papers reported here is cited as Discussion Paper Series No. 0702 

The second is cited as Paper 0703

Full background on legislation and official figures on delayed discharge

Further discussion and research is contained in the following refereed articles:

Henwood M (2006), “Effective partnership working: a case study of hospital discharge”, Health and Social Care in the Community, 14(5), pp400-407

Baumann M et al (2007), “Organisation and features of hospital, intermediate care and social services in English sites with low rates of delayed discharge”, Health and Social Care in the Community, 15(4), pp295-305


The need for improved data collection

Some of the implications concern the need for better data collection practice, others concern how we conceptualise the problem of delayed discharge. On the former, the need is to extend the data collecting system to cover the full spectrum of hospital, intermediate, residential and community-based services, and to ensure that accelerated hospital discharges are not compromising the long-term and overall quality of care. On the latter, the government has been too hasty in simply assuming adult social care is the guilty party on this matter and therefore needs to be financially punished.

Benefits of a whole-system approach

The fact is that a high level of delayed discharges is a symptom of system malfunctioning rather than a cause. The real task is for local stakeholders to undertake a system-wide audit of why delayed discharge is happening and how, using a whole-system approach, it can be addressed. Unless and until this happens, agencies will be locked into interminable controversies about funding responsibility and organisational culpability, practitioners will be pressed into taking actions that suit financial imperatives rather than individual needs, and patients will be unwitting victims.

benefits of a whole-system approach

More from Community Care

Comments are closed.