Winter Pressures in NHS Scotland 2008-2009

Winter Pressures in NHS Scotland 2008-2009

Report for the Emergency Access Delivery Team, Scottish Government

Dr Daniel Beckett

CONTENTS

CONTENTS

FIGURES

TABLES

Winter Pressures Report Executive Summary

Introduction

Methods

Assessment of pressure

Analysis of qualitative data

The winter planning process

NHS24

Scottish Ambulance Service

Out of hours General Practitioner Services

Hospital admissions

Capacity and demand within secondary care

Hospital discharges

Analysis of quantitative data

Long term trends

Winter 2008-2009

The 4 Hour Standard for access to emergency care

Patient characteristics

Special Health Board data

GP out of hours services

Summary, Conclusions and Recommendations

Summary

Conclusions

Recommendations

Appendix 1: Questionnaire

Appendix 2: List of interviewees

Appendix 3: Examples of good practice

Appendix 4: Festive monies

FIGURES

Figure 1. Mean temperature anomaly winter 2008-2009......

Figure 2. Winter comparison of mean temperatures for Scotland 2006-2007 & 2008-2009

Figure 3. Days of air frost anomaly winter 2008-2009

Figure 4. A&E attendances per day vs proportion admitted - Dec 2008-Jan 2009-reduced staffing

Figure 5.A&E attendances per day vs proportion admitted - Dec 2008-Jan 2009-full staffing

Figure 6. Emergency department attendances and admissions NHS GG&C winter 2008-2009

Figure 7.Emergency department time profile Dec 2008 – good performance

Figure 8. Emergency department time profile dec 2008 – poor performance

Figure 9. Beds occupied by patients admitted as emergencies (weekly average)

Figure 10. Bed occupancy levels for core and non-core sites – Apr 2008-Mar 2009

Figure 11. Delayed discharges - Scotland - Jan 2006-Apr 2009

Figure 12. First A&E outpatient attendances, Scotland, 1998/99 – 2008/09

Figure 13. Emergency department admissions vs all emergency admissions Feb 2008-Jan 2009

Figure 14. Attendances, admissions & transfers from core EDs winter 2008-2009

Figure 15. Scotland emergency & elective inpatient admissions winter 2008-2009

Figure 16. Scotland emergency & elective inpatient admissions inc. daycases winter 2008-2009

Figure 17 admission/discharge profile with 4 hour breaches from core sites dec 2008-jan 2009

Figure 18. Number of inpatient discharges per wk dec 2008-jan 2009

Figure 19. Percentage of A&E attendances meeting 4 hour Standard, July 2007 to April 2009

Figure 20. Reasons for 4 hour breaches, Dec 2008-Jan 2009

Figure 21. Number of 12 hour breaches, June 07 to April 09

Figure 22. Emergency admissions per age group, February 08-February 09

Figure 23. Emergency admissions winter 2008-2009 split by respiratory/non-resp diagnosis

Figure 24. Emergency admissions winter 2006-2007 split by respiratory/non-resp diagnosis

Figure 25. Scotland level monthly emergency admissions for all diagnoses Feb 08-Jan 09

Figure 26. NHS24 call demand Winters 2006-2007 & 2008-2009 and summer 2008

Figure 27. NHS24 winter 08/09 total call demand

Figure 28 SAS Category A incidents by week, mainland Scotland

Figure 29. SAS Category A performance mainland by subdivision, Dec 06/Jan 07/Dec 08/Jan 09

Figure 30. GP out of hours activity

Figure 31. GP out of hours performance for 1 hour home visits

TABLES

Table 1. Mean scores given on Likert questions by Clinicians, Management and GPOOH staff…......

Table 2. NHS Scotland emergency and elective admissions Dec 2008 and Jan 2009 compared with the five year monthly mean

Winter Pressures Report Executive Summary

Feedback from NHS Health Boards

BACKGROUND

1. In March 2009 the Scottish Government Emergency Access Delivery Team commissioned a report to review the pressures experienced, and response by NHS Scotland during winter 2008-2009.

2. There has been a significant improvement in the 4 hour emergency access waiting times, increasing from 87% in June 2006 to delivery of the current 98% 4 hour HEAT Standard. However, there was a drop in performance in 11 out of the 14 territorial Health Boards against the 4 hour Standard in December 2008 and January 2009. Overall performance for NHS Scotland during these months dipped below 98% (96.7% and 96.5% respectively). Additionally, there was media interest in how well the NHS had handled winter in parts of Scotland, particularly the central belt, with articles about long trolley waits, and hospitals not coping with adverse weather conditions. Recent figures (March 2009) show performance improved to 97.7% with 140,000 attendances (compared with 128,084 in December).

3. The review was undertaken by Dr Daniel Beckett, Acting Consultant in Acute Medicine at the Royal Infirmary of Edinburgh, with support from NSS Information Services Department (ISD) and Scottish Government Analytical Services Directorate (ASD). Each NHS Health Board in Scotland was visited, plus the relevant Special Health Boards, to obtain a wide range of professional and staff perceptions about their local experience of winter using semi-structured one-to-one interviews. Quantitative data and information was provided by ISD and ASD.

4. The report focused upon December and January as these were the months with the poorest performance against the 4 hour Standard. Comparison was made with previous winter performances, particularly 2006-2007 (as 2007-2008 was widely considered to have been atypical)

KEY FINDINGS

5. The key findings from the qualitative and quantitative aspects of this report are summarised below.

Hospital Admissions

6. There was a perception amongst Health Boards that winter 2008-2009 had been ‘busier than previous years’ in terms of total numbers of admissions and that the peak had commenced earlier. Health Boards commented that the age profile of patients admitted over winter appeared to be older, with more patients suffering from respiratory disease resulting in greater lengths of stay. Data from the SMR01 dataset confirmed that, compared with the five year monthly mean, there was a 7.9% increase in all emergency admissions across the NHS in Scotland in December 2008, followed by a 1.8% increase in January 2009 (Figure 16, main report (p46)). Furthermore, there was an 11% increase in the number of patients admitted to hospital with respiratory illness over December and January compared with winter 2006-2007. However in practice this equates to only 2 extra respiratory patients per day across NHS Scotland. There was no evidence for a disproportionate increase in admission of elderly patients or greater length of stay.

Hospital Discharges

7. Low levels of hospital discharges, particularly over the festive period, were highlighted as a cause for concern by most NHS Health Boards over winter 2008-2009.

8.Figure 18, main report (p48), shows admission/discharge profiles across NHS Scotland (from core sites) plus number of four hour breaches. There is a consistent admission/discharge profile in December, with peaks of admissions at the start of the week, mirrored by a peak of discharges towards the end of the week. Admissions outnumbered discharges every weekend (and on Mondays) with a surge of discharges on Christmas Eve. This was followed by the 11 day holiday period, and for nine of these 11 days, admissions outnumbered discharges. The net effect was that hospitals had high levels of bed occupancy when the elective programme restarted at full capacity on 5 January, resulting in a spike in four hour breaches. The system attempted to return to the previous admission/discharge profile over the following week, but had not recovered by the following Monday, 12 January, and further four hour breaches were noted.

9. Reasons highlighted for this reduction in discharges over the festive period included:

• lack of consultant staff in downstream wards;
• lack of discharge infrastructure over the festive period (e.g. Patient Transport Services, Allied Health Professionals and social work); and
• perceived lack of coordination of decision making in the system over the festive period.

10. Tools for discharge planning, such as Estimated Date of Discharge (EDD), and Nurse Led Discharge (NLD) were used patchily, and in some Health Boards not at all.

11. Health Boards reported almost universal improvement in numbers of delayed discharges, with NHS Scotland achieving zero delayed discharges over 6 weeks by April 2008 and 2009 (although this was not achieved in every month). A small number of Health Boards continued to be challenged by significant numbers of delayed discharges over 6 weeks.

Capacity and Demand Planning

12. Nine of the 14 territorial Health Boards have developed an internal tool to predict unscheduled activity, and these were largely found to be accurate. Two Health Boards employed tools to predict discharges based on previous discharge patterns. Despite Health Boards being encouraged to use System Watch, there was little use for medium to long term predictions of activity despite its proven accuracy. Figure 9, main report (p31), shows the System Watch prediction for winter 2008-2009. Activity started to increase early (2 November), but then short term prediction followed well after 2-3 weeks.

13. Eight out of 11 mainland Health Boards opened additional capacity beds in their acute sites this winter. Many Health Boards had difficulties accessing the full complement of community beds, despite the acute site being near, or over-capacity. These difficulties included:

• lack of Patient Transport Services;
• complex referral pathways; and
• patient choice.

Escalation Plans

14. There was variation between Health Boards in the effectiveness of local escalation plans. Most Health Boards had a bed management escalation plan, but the triggers for escalation varied between predicted activity, actual observed activity, or simply perception of activity. There were reports of managers and clinicians (including primary care) becoming desensitised to red alert. Conversely there were reports of middle management being reluctant to escalate, or senior managers refusing to escalate to red alert.

15. CHP involvement in escalation plans was variable, with one example of an escalation plan being developed by CHP senior management without clear involvement of CHP middle managers, who were not fully sighted on this and were unable to respond when necessary.

Elective Activity

16. Eight of the 14 territorial Health Boards continued with elective work until Christmas Eve and between Christmas and New Year, whereas six Health Boards ran a ‘cancer and urgent only’ service over the festive period. The decrease in elective admissions on 29 December (60% of a ‘normal’ Monday) is shown on Figure 16, main report(p46).

17. The perception in several Health Board areas was that the introduction of the 18 week Referral to Treatment Target, and the disbanding of the Unscheduled Care Collaborative, de-prioritised the 4 hour Standard. Examples of this included:

• surgical wards (with staff available) remaining closed over the festive period despite eight hour, or greater, trolley waits in the Emergency Department; and

• waiting list initiatives on 5 January despite clear predictions of high levels of unscheduled medical activity.

18. Seven Health Boards did not cancel any elective procedures due to lack of beds, five cancelled a small number (<15 each) and two cancelled significantly more.

Staffing

19. Three major staffing challenges over winter were highlighted:

• Potential problem with non-clinical staff retention in out of hours GP services because staff employed under Agenda for Change were not paid extra for working unsociable hours over the festive period.
• Implementation of Modernising Medical Careers (MMC) and nationalised medical recruitment has led to a number of medical posts, particularly within acute specialities, remaining unfilled.
• Many sites did not have sufficient consultants in the hospital to deal with the predicted activity over the two week festive period and to facilitate quicker discharges in down stream wards.

• There was a perception of lack of social work availability due to significant amounts of annual leave being taken over the festive period.

Boarding Patients (outliers)

20. Each of Scotland’s 11 mainland Health Boards use boarding of patients outside their own speciality beds as a solution to capacity issues. Over winter 2008-2009 in some sites up to 60% of all medical patients were boarders, occupying more than 10% of the total bed complement. There has been a recent move to board patients from the Admission Unit (and in exceptional circumstances the Emergency Department) before initial consultant review. This should be considered a clinical governance issue.

Recommendations

These recommendations should be considered alongside the many examples of good practice detailed in Appendix 3 to this report.

• Health Boards should ensure that their winter planning starts early and that the process includes Community Health Partnerships and Social Work Departments. There should be a clear relationship between the winter plan and pandemic ‘flu plan.

• Integral to the winter plan should be the escalation plan. This should involve all stakeholders including Community Health Partnerships. This includes the utilisation of beds in Community Hospitals, and protocols for referral and transfer should be agreed to resolve issues relating to perceived bed ownership.

• System Watch should be used systematically for long to medium term predictions of unscheduled activity, and predictions acted upon to create the necessary capacity, in terms of beds and to support initiatives to reduce admissions. Consideration should also be given to the use of System Watch for planning of elective activity over the winter months.

• Daily bed meetings should take place at every site, and should occur twice daily during the winter period. Consultant medical staff should have greater awareness of bed management issues, including escalation plans for sites.

• Health Boards should undertake more accurate modelling over the festive period to plan elective capacity and optimise the use of bed capacity, including maximising the bed capacity in community hospitals. This may then enable hospitals to reduce the number of elective admissions on the first Monday in January. Further consideration should be given to front loading the first week in January with minor procedures, and back loading with majors. Also medical elective activity (such as clinics and endoscopy lists) could be back loaded during this week.

• Medical Directors should ensure that appropriate numbers of consultant medical staff are on site to deal with the predicted activity over the two week festive period.

• Health Boards should aim to eliminate boarding of patients as a solution to bed capacity problems. Specifically, the boarding of patients from the Acute Medical Unit and/or Emergency Department should not occur (this includes ‘treat and transfer’ policies, with the exception of tertiary care referrals).

• The level of discharges over the holiday period should be improved. This might include:
• increased consultant presence with dedicated discharge ward rounds in downstream wards;
• utilisation of a rapid response team (or equivalent) of AHPs with access to homecare packages without recourse to social work assessment; and
• re-energising and establishing ownership of the Estimated Date of Discharge policy, plus introducing Nurse Led Discharges (NLDs).
• Patients should be discharged early in the day, as this is key to maintaining capacity. Planning of discharge ward rounds should take this into account.

If all the above measures have been undertaken, including consultant review and discharge of downstream patients, and all capacity beds filled (including community beds) then the 98% standard for emergency access of care should be achievable. Health Boards should note that if there are ongoing difficulties then priority should be given to emergency admissions over routine elective procedures.The Scottish Government has, for the last 10 years, made it clear that clinical decision making always trumps routine elective targets.

Introduction

1. The National Emergency Access Delivery Team (EADT) works closely with Scottish Government Health Directorates to provide direction and support to NHS Health Boards to:

• deliver local improvement trajectories for reducing rates of attendances at the Emergency Department (HEAT target T10);
• monitor Health Boards’ performance in relation to the maximum 4 hour wait; and
• improve whole systems winter planning.

1. In March 2009 the EADT commissioned a short term review of pressures experienced and response by NHS Scotland over winter 2008-2009. The main driver for this was the dip in performance against the 4 hour Standard for emergency access in December 2008 and January 2009 in some parts of Scotland. Additionally, there had been unfavourable media interest regarding NHS handling of winter pressures, particularly in the central belt, with articles about extended trolley waits, and hospitals unable to cope with the adverse weather

1. The purpose of this review was to:

-provide a description of unscheduled care systems in Scotland over winter, including levels of activity and pressure points from November 2008 – March 2009 (this being defined as the ‘winter period’);

-assess the extent to which the system ‘coped’ or showed signs of strain;

-describe the winter planning response, including what worked and what didn’t;

-derive lessons for the future, and explore how recommendations may be implemented; and

-identify the extent to which community/primary care can improve the effectiveness of the whole system of unscheduled care.

1. The review was carried out by Dr Daniel Beckett, Acting Consultant in Acute Medicine at the Royal Infirmary of Edinburgh, with support from NSS Information Services Department and the Scottish Government Analytical Services Directorate. Each NHS Health Board in Scotland was visited, plus the relevant Special Health Boards, to obtain the perceptions of staff using semi-structured one-to-one interviews. Quantitative data and information was provided by ISD and ASD. The Key Learning Points and examples of good practice will be shared with the Service, both at the National Winter Planning Conference (June 2009) and through distribution of this report, in order to inform planning for winter 2009-2010.

Methods

Qualitative data

1. All 14 territorial Health Boards across NHS Scotland were visited, plus the relevant Special Health Boards. Seventy interviews, using a semi-structured questionnaire, were undertaken over a 10 week period. All interviews were digitally recorded and transcribed by the author. All data was anonymised following transcription. A minimum of two people from each Health Board were interviewed. These included:

• Hospital management, including Chief Executives, Directors of Operations, General Managers and Senior Bed Managers;
• Secondary care clinical staff, with the focus on clinical leads for Emergency Medicine and Acute Medicine;
• Clinical and management staff from GP out of hours services;
• Social work representation; and
• Mental health representation.

The Special Health Boards visited were: