A recent NHS study of hospital readmissions recognised the contribution of social care but is too clinically focused
Title: Emergency Readmission Rates: further analysis. Authors: Panos Zerdevas (NHS Finance and Investment Directorate) and Charles Dobson (NHS Medical Directorate)
The research explores and updates data on emergency readmission rates to acute hospitals in England. The data show that emergency readmissions have been rising for some time and the paper seeks to explain this trend. While some of the explanation lies in the types of cases being treated and in some developments in medical and clinical practice, the role of social care and community services is also critical, especially in the case of older people with a long-term condition.
The use and cost of acute hospital beds has become a key health policy issue in recent years, initially with an emphasis on minimising delayed discharge and more recently with a focus on reducing avoidable admissions. Particular attention is now being paid to cases of complex long-term conditions, typically older people, who are the most frequent users of unplanned secondary care. It is against this background that research from the Department of Health seeks to shed new light on those people who continue to return on an unplanned basis through emergency readmissions.
Building on earlier work by the National Centre for Health Outcomes Development (NCHOD), the analysis examines the rise in readmission rates in recent years, attempting to discern underlying trends and break down the data to a sufficient level of detail to offer some tentative interpretations. In particular, unlike the NCHOD analysis, data are split into two groups – those aged 16-74 and 75-plus. The three data sources used are: existing NCHOD data on readmission covering the years 1998-9 to 2005-6, newly analysed data for 2006-7, and reactions to initial findings from a workshop of clinicians.
For the purposes of the research, ”emergency readmission” is defined as any emergency admission into hospital within 28 days or fewer following discharge from a previous stay in hospital. The indicators exclude day cases, maternity patients and those with mental health problems or cancer, where such readmissions are more likely to be expected. Given this definition several key findings emerge that are relevant for social care and community services.
Rises in readmission rates: There has been a steady rise in readmission rates in the decade since 1998, though there is some recent evidence of stabilisation. The readmission rate for those aged 16-74 has increased over the period from 6.9% to 9.1%, and for those aged 75-plus from 10% to 13.9%. Older people have higher readmission rates than younger people because they are more likely to be frail or more likely to suffer from long-term conditions associated with relatively high rates of readmission. For older people alone the number of emergency readmissions has gone up from 88,000 in 1998 to 149,000 in 2006, an average of 1,000 per adult social care authority.
The impact of length of initial stay: It would be tempting to surmise that part of the explanation for rising readmissions is the impact of shorter initial stays as hospitals seek to release beds quickly for new cases, but the study finds little statistical evidence to support this hypothesis. Clinical discussion at the workshop confirmed the view that equating the increasing rate of readmissions to reductions in the quality of hospital care was too simplistic.
Changes in the case mix of admitted patients: It is possible that the increases in day cases (which are excluded from the definition of the readmission rate) and better support in community settings mean that patients admitted in the initial episode are more likely to have relatively severe problems than 10 years ago. The study estimates that 25% of the increase can be explained by this factor, but that further analysis is needed.
Age-band differences: As noted above, it is older people aged 75-plus who are driving up readmission rates and staying for longer once readmitted, especially where they have multiple long-term conditions. The main conditions are shown to be heart disease and respiratory conditions such as chronic obstructive pulmonary disease and asthma.
What are we to make of these findings and what are the implications for social care? Perhaps the most obvious remark to make is that this research is very clinical in nature.The paper emanates from the NHS Medical Directorate and the 11-strong workshop used as a sounding board to discuss the findings did not contain one member with a background in social care or even local government more widely. With no contribution from social care, and no analysis of how a more integrated approach might help to address the readmission dilemma, the research has come out of the health silo.
But it would be wrong to suggest that the paper shows no awareness of the potential contribution of social care – indeed at several points the availability (or otherwise) of effective community services is identified as one possible reason for the variations in local readmission rates.
There is some passing praise for the Partnership for Older People Project (Popps) pilots, but for the most part the clinicians consulted as part of the study seem to equate community care with the activities of primary care trusts rather than local councils.
The most common research and policy response to the problem of unplanned hospital admissions and readmissions has been to redouble the search for “predictive tools” that can be used to identify those individuals most at risk of hospitalisation. Perhaps the best known is the Patient At Risk of Rehospitalisation (Parr) case-finding tool commissioned by the Department of Health – a hospital-based approach focused on those who have had an admission and using rehospitalisation as a trigger. Others such as the slender Emergency Admission Risk Livelihood Index (Earli), developed in North West England, try to be applicable to both clinical and community settings.
The usefulness of such tools is limited by two factors. First, the lack of evidence that targeted intervention on people identified by this process (such as community matrons using a care management approach) is effective in reducing admissions. Caution needs to be exercised in attributing reductions in admission rates in high-risk groups to such interventions for this could equally be caused by relatively high mortality and regression towards the mean in surviving patients. What is needed here is a tool that helps to predict the risk of needing intensive social care rather than simply hospital readmission.
Second, and more importantly, these clinically led approaches fail to appreciate the need to involve the whole system including social care, looking at readmission as an “upstream” and not just a “downstream” issue. In such an approach it will be necessary to go beyond clinically targeted interventions to encompass self-care, self-directed support (for users and carers), prevention strategies (covering the use of social capital, the role of social housing and social care) and getting GPs to work differently with colleagues in a range of community settings.
This is precisely the intention of the Department of Health’s Integrated Care Pilot Programme which is just getting under way, but even here there is an assumption that the innovations will be clinically led.
Bob Hudson is visiting professor of public policy, School of Applied Social Sciences, Durham University
This article is published in the 4 December 2008 edition of Community Care magazine under the headline Emergency readmission rates: everybody’s business