As the NHS looks for new ways of helping patients to avoid the need for unplanned hospital admission, researchers at the Nuffield Trust have conducted an evaluation of an innovative scheme called Virtual Wards, which has the potential to prevent readmissions using predictive modelling.

Many NHS organisations use computer models (called ‘predictive models’) to help them identify individuals who are at high-risk of future hospital admissions, so that these people can be offered extra support to keep them healthier and hopefully avoid the need for emergency admissions.

In certain parts of the NHS in England, such people are being offered extra support through a system of ‘Virtual Wards’. These aim to prevent unplanned admissions by using the systems of a hospital ward to provide multidisciplinary case management in the community.

The Virtual Wards work just like a hospital ward, using the same staffing, systems and daily routines, except that the people being cared for stay in their own homes throughout.

This project looked at some of the early adopters of the model in England, and tracked patterns of use of health and social care services for patients on a Virtual Ward

With funding from the National Institute for Health Research Service Delivery and Organisation Programme, researchers at the Nuffield Trust linked anonymised data to assess health and social care activity and the associated service costs of the Virtual Wards in Croydon, Devon and Wandsworth.

We used pseudonymous data (data from which patients’ identities have been removed to ensure confidentiality) to analyse patterns of hospital, GP and social care use in the three areas. We identified all those individuals who received care on a Virtual Ward, and tracked their use of health and social care services over time.

We then compared the hospital use of the people who received Virtual Ward care with similar patients (called ‘comparator’ groups) drawn from national and local data to see whether there were any differences.

This research also aimed to find out how much it costs to run a Virtual Ward, and what effect – if any – they have on the use of health and social care services.

The ‘economic analysis’ involved working closely with the Virtual Ward staff, their managers and finance officers, to understand how Virtual Wards work in practice. Administrative data, interviews and diaries were used to calculate the costs of setting up and running a Virtual Ward.

Once we gathered all of this information, we pulled it together to work out the overall costs or savings for the NHS and local authorities, and this was used to develop a method by which councils and clinical commissioning groups (CCGs) can choose between different arrangements of Virtual Wards locally.


We found that:

  • Each of the study sites had implemented variants on the Virtual Ward and in Croydon, by far the largest site, multidisciplinary preventive care was only offered during a short initial pilot period before changing to case management by community matrons. Our findings are therefore predominantly related to patients who received ‘standard’ case management rather than management from a Virtual Ward team.
  • Compared with matched controls, we found no evidence of a reduction in emergency hospital admissions for patients who received this type of care in the six months after starting the intervention. Nor did we find evidence of a reduction in ambulatory care sensitive hospital admissions or mortality in this period. We did, however, observe a reduction in elective hospital admissions and in outpatient attendances in the six months after starting the intervention.
  • The direct costs of the interventions varied considerably between the three sites, ranging from approximately £3 per patient per day in Croydon and Devon, to £17 per patient per day in Wandsworth, reflecting the heterogeneous nature of the interventions being studied. Over the six month follow-up period of analysis these direct costs were of the order of £510-£2,890 per patient.

We suggest that health service commissioners and providers should consider a number of factors. Firstly is the question of what it means to be a Virtual Ward. There were different 'forms' of Virtual Ward in this study and we suspect an even wider number of variants in other settings.

Second, our descriptive analyses showed that many of the patients who received the intervention had highly complex illnesses that have important implications for a local health and social care economy.

Such patients groups represent an important challenge not simply in terms of the costs of care but also in terms of the quality of care provided. Our study was not designed to assess changes in the quality of life of Virtual Ward, but there are anecdotal reports that patients were highly satisfied with the intervention.

Thirdly, Virtual Wards are part of a generic approach to long-term care which may be justified on other terms, for example as ways to improve the quality of communication between community health staff, the continuity of care, patient experience or safety.

Read the full findings of our research in a report by the National Institute for Health Research Service Delivery and Organisation Programme and an article for the International Journal of Integrated Care both published in November 2013.

We are currently working with Devon to undertake an evaluation of much larger groups of patients from that Virtual Ward, as there were so few cases from Devon in our original analysis. We hope that this work will be completed by Spring 2014 – subject to agreement on access to information.

This project was funded by the National Institute for Health Research Service Delivery and Organisation Programme (project number 09/1816/1021).

The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the HS&DR Programme, NIHR, NHS or the Department of Health.

Email to a friend

Your message will be:

I thought you might be interested in this page on The Nuffield Trust website.