Reducing emergency admissions Contents

2Understanding what works

The evidence base

12.Providing cost-effective alternatives to emergency care needs input from across the health and social care system. NHS England, and its partners, have set up several programmes which aim to reduce emergency admissions through working with different parts of the care system: The urgent and emergency care programme aims to improve emergency and urgent care and ease the pressure on the emergency system; the Better Care Fund aims to integrate health and social care, the new care models aim to integrate primary and hospital care; and NHS RightCare and Getting it Right First Time try to help local areas understand how their performance compares with other similar places.22

13.NHS England could point to some indicators of success in these programmes, for example, the new care models showing, on average, a slowdown of growth in emergency admission rates.23 However, NHS England has not been able to unpick what particular interventions in these programmes work best to reduce emergency admissions. In fact, the interventions are not always based on what works well in practice. The evidence base for these interventions are mixed and, in some cases, quite poor.24 Without an understanding of what works, NHS England cannot target its efforts to improve and get the most effective use of taxpayers’ money.25

14.Factors such as deprivation and demographics can have a major impact on emergency admissions rates. Even when these factors are taken into account at local level, the number of emergency admissions in England in 2016–17 varied considerably, between 73 and 155 admissions per 1,000 people.26 NHS England told us about a programme called Getting it Right First Time which is a clinically-led programme that looks at variation within hospitals and began by looking at surgery. NHS England believes that it will contribute to the identification of clinical best practice within hospitals.27 However, we were unconvinced that NHS England was drilling down enough to find out the real reasons for these differences in the populations, and sharing this understanding with NHS teams.28 When challenged on the lack of evidence of the impact of particular interventions, NHS England does not seem to understand which particular interventions are working or why. Without a good understanding of what causes these variations, NHS England and its partners do not know whether local social and health care practices are causing different rates of emergency admissions than elsewhere.29

Poor data

15.In most cases, a shorter stay in hospital is best for people, particularly the elderly, as they lose mobility quickly if they do not keep active, and their ability to do everyday activities can reduce quickly while in hospital.30 NHS England and NHS Improvement are trying to promote a model of emergency care, known as daycase or ambulatory emergency care, in which people are admitted to hospital but do not stay overnight. NHS England explained that this model of care can take many forms including a specific “ambulatory care” facility on the hospital emergency floor or a team that specialises in frail patients.31 It considers that this type of care is one of the positive steps it is taking to manage patients closer to their home and in a way that is better for patients.32

16.However, hospitals do not record daycase emergency care consistently. Some hospitals record these patients as an emergency admission while others record them as outpatients.33 There is no guidance as to how hospitals should record these patients and NHS England acknowledged that the system was not set up to enable hospitals to record this care consistently.34 This inconsistency creates two problems. First, it prevents NHS England knowing how successful its efforts are in providing what it considers better care for certain patients, and whether patients are being spared an unnecessary overnight stay in hospital. Second, it carries the risk that hospitals may game the data to get higher payments through the tariff system, which pays hospitals more for emergency admissions than for outpatients.35 We asked NHS England about the danger of gaming the system. It responded that clinicians are rarely aware of how the tariff works and would treat the patient in the best way possible. It accepted that the data needed improving and pointed to a pilot it has started in six areas to improve data on daycase emergency care.36

17.There are similar problems with the data on the numbers of people being readmitted to hospital. Readmissions can indicate the success of the NHS in helping people to recover from illness or injury. They can happen for many reasons and may not always be preventable.37 NHS England told us that clearly there have been occasions when people have been discharged from hospital too soon, or where the community and social care they needed was not in place as expected.38 However, NHS England explained that readmissions are not always a bad thing and may result from a push to get people out of hospital as quickly as possible given the health problems associated with long stays in hospitals.39

18.However, the NHS does not record data on readmissions and so is unaware if readmission rates are approaching levels that could be harmful, and be an indicator of failures in care. We challenged NHS England to be clearer about when a readmission was a positive indicator or a result of poor judgement, but there is evidently more work needed to give this clarification.40 A report by Healthwatch England in October 2017 on data from 72 trusts estimated that readmission rates have risen as much as 22.8% between 2012–13 and 2016–17.41

22 C&AG’s Report, paras 2.3–2.9

23 Qq 50, 51

24 C&AG’s Report, para 3.26

25 Qq 43–47

26 C&AG’s Report, para 3.24

27 Q 68

28 Qq 68–69

29 Qq 62–69; C&AG’s Report, para 3.24

30 Q 18; C&AG’s Report, para 1.7

31 Q 13

32 Q 12

33 Qq 13–16; C&AG’s Report, paras 13, 3.32

34 Qq 13–16; C&AG’s Report, para 3.32

35 Qq 12–16

36 Q 16

37 C&AG’s Report, para 3.3

38 Q 18

39 Qq 17–19

40 Q 26

41 C&AG’s Report, para 3.4

Published: 8 June 2018