Agenda Item No: 8 PURPOSE: To highlight to the Board the importance of reducing avoidable mortality and to propose a series of evidence based measures that could significantly improve patient safety. IMPLICATIONS: Objectives to which issue relates Risk issues Objective 8: To ensure compliance with the statutory requirement for quality and the delivery of safe, high quality patient care within a reporting and learning culture Failure to analyse patient safety reports and action any associated weaknesses in patient care could expose patients to risk in the future, compromise the s reputation. Financial HR Extra investment in reducing mortality and improving patient safety is highly likely to bring a return in terms of extra lives saved. Support will be required from a statistician, data information analysists and audit staff Healthcare/ National Policy The is committed to enhancing the quality of the patients experience, their treatment and care. Legal issues None RECOMMENDATIONS: The Board is asked to approve the contents of the paper. DIRECTOR: PRESENTED BY: AUTHORS: Medical Director Medical Director Medical Director & Transitional Manager, Strategy DATE: 28th May 2008 1
1. Introduction This has an ambition to be a high performing organisation delivering excellence in healthcare. As part of this process the board has determined to put patient safety at the forefront of its business to be open with the public on key clinical performance data. There are already a raft of important patient safety initiatives within the which are delivering improvements in patient outcome. The purpose of this paper is to highlight the importance to the of focussing on reducing avoidable mortality as a means to improving patient safety and proposes a series of evidence based measures to achieve this. Avoidable mortality can be defined as deaths that should not occur given current medical knowledge and technology 1. There is wide variation in the level of avoidable mortality in NHS s in England. 2 The proposals suggested in this paper are about using avoidable deaths as a focus for addressing possible systemic problems in the provision of clinical care. Hospital specific mortality rates are becoming publicly available and will increasingly be used as an indicator of the quality of care. Reducing avoidable mortality will give patients more confidence in their care and clinical outcomes, assure staff at all levels of the safety of the care they provide and promote a reputation of being a safe. 2. Background A significant amount of work has recently been carried out reviewing the s mortality rates and findings were reported to the Board in closed session in March 2008. In summary, the review found: No indication of mortality rates being inflated due to inadequate clinical care. Primary diagnostic coding is, in the main, accurate and therefore the correct HRG s are being allocated. The s crude mortality rate of 2.15% for the calendar year 2007 is average when compared against the national rate for deaths in hospital in the UK. When E&N Herts s mortality for Q1&2 2007 is compared to the CHKS total database (covering 140 hospitals), the is performing at an average rate in Medicine and Surgery and better than average in Women and Children s. The Medical Director was satisfied that the has mortality rates within a normal range when compared with other s. Hospital Standardised Mortality Ratio (HSMR) indicates death rates for patients having certain types of procedures. The ratio compares the actual number of deaths with the expected number of deaths, and takes account of factors including age, sex, diagnosis, whether the admission was planned or emergency, and the length of stay. This is the methodology used by Dr Foster in its nationally published reviews of hospital mortality but only includes 80 three character ICD diagnosis codes, which are mostly surgical. An HSMR of 100, means that the number of patients who died is exactly as it would be expected taking into account the standardisation factors. An SMR above 100 means more patients died than would be expected; one below 100 means that fewer than expected died. East and North Hertfordshire NHS s HSMR for 2005/6 was 98%. The figure for 2006/7 is due to be published by Dr Foster in June. 1 French, K, and Jones, K (2006) Impact of definition on the study of avoidable mortality: geographical trends in British deaths 1981 1998 using Charlton and Holland s definitions in Social science and medicine 62(6): 1443 56 2 Jarman, B, et al, Explaining differences in English hospital death rates using routinely collected data BMJ, 318:1515 1520, 1999 2
Nationally and internationally there is evidence of decreasing mortality (improving medical care) over the past few years. Figure 1 shows the English HSMR, using the year 2000/01 as reference, over the past twelve years. Figure 1 English HSMRs using year 2000 (Apr 2000 Mar 2001) as reference 120 115 110 105 HSMRs (95% CIs) 100 95 90 85 80 75 70 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Ref: Professor Sir Brian Jarman, Reducing Avoidable Deaths in Hospital, April 2008 There is a substantial amount of published scientific evidence on hospital mortality rates and the factors, which affect them. There are a variety of strategies that the could employ to reduce hospital mortality. In broad terms they are: a) Avoidance or optimisation of treatment of high risk patients. Whilst the has no intention of denying potentially effective treatment to any patient, there are now tools available which provide accurate risk stratification on a patient by patient basis and such information could be used by both patients and their clinicians in making an informed decision as to whether to proceed with a high risk procedure or treatment. This information can also be used to identify patients at high risk and ensure optimal peri operative care. b) Reducing risks of medical accidents. The already has established processes in place to minimise the risk of medical accidents. There are, however, areas where further improvements could be made. c) Improving the quality of care given to seriously ill or deteriorating patients. Building reliable systems with standardised interventions to rescue the deteriorating patient has been shown to impact significantly on reducing hospital mortality rates. 3
3. What has been done: Significant steps have already been taken within the to improve patient safety and to reduce avoidable mortality. These include: 3.1 Venous thromboembolism prophylaxis The has implemented venous thromboembolism prophylaxis (which reduces the risk of deep vein thrombosis and pulmonary embolism) in line with the National Institute for Health and Clinical Excellence guidelines 3. 3.2 Transfusion Committee The transfusion of blood and blood products carries a small but significant risk of serious adverse outcome. The has constituted a blood transfusion committee which oversees protocols and clinical developments in this area. The focus of the transfusion committee for 2008 is to enable a reduction in the use of blood and blood products, where clinically indicated, and to oversee the roll out of NPSA prescribed competency assessments for all staff involved in transfusion. 3.3 Enhanced Recovery Programme The is in the process of implementing an Enhanced Recovery Programme. Over 20,000 patients die nationally each year following surgery 4 Of those patients who die following surgery, the vast majority have other serious medical conditions at the time of surgery, although data from the National Confidential Enquiry into Patient Outcome and Deaths suggest that patients actual risk of death is not being fully appreciated prior to their surgery. The report suggests three interlinked areas of surgical care and decision making that will reduce peri operative death: o improved pre operative assessment, triage and preparation: objective evaluation prior to planned major surgery can identify patients with increased risk profile, and for some, this may result in a decision not to proceed with major surgery. This is can effectively be achieved by pre operative cardio pulmonary exercise testing. The has introduced CPX equipment. This comprises of an exercise bike and an electrocardiograph to assess heart and lung function to the point of anaerobic threshold. Its results correlate well with post operative survival. o Improved intra operative care: improvements in fluid administration and other interventions significantly reduce both rates of post operative complications and mortality, as well as significantly reducing both the length of hospital stay and the overall number of ICU bed days used. o Improved use of post operative resources: the planned transfer of patients to HDU or ICU, or to receive outreach critical care, can lead to improved post operative outcomes and can reduce the overall number of total bed days used as well. These three elements form the Enhanced Recovery Programme, and the improvements will help patient outcomes by reducing the number and severity of complications suffered following major surgery as well as decreasing the mortality rate. 3 Venous thromboembolism, NICE clinical guideline 46, National Institute for Health and Clinical Excellence, April 2007. 4 Modernising Care for Patients Undergoing Major Surgery. (Based on a report by the Improving Surgical Outcomes Group in June 2005). 4
3.4 Trauma A Trauma Team has been developed within the. The presence of orthopaedic surgeons on call has been increased. The is in the process of joining the Trauma Audit and Research Network to improve mortality rates. A computerized trauma tracker is being introduced to improve the speed and process of care for trauma patients. More operating time has been dedicated to trauma patients to ensure that they can be operated on within the first 24hours of making the decision to operate. This remains challenging. Links have been greatly improved with Medicine for the Elderly and plans are under consideration to further develop these links by having an orthogeriatric model of care. The aim of this is that all fractured necks of femur over 65 will be under the care of the Care Of the Elderly team from admission, with input from the surgeons provided up until 72 hours post op. 3.5 Pre operative briefings The Royal College of Surgeons recommends holding a multidisciplinary team meeting attended by the surgeon, the anaesthetist and a member of the theatre staff to discuss each patient, the operations and equipment requirements before the start of each operating list. Between each surgical case they recommend a pre operative pause to enable time to check each patient, the correct operation and surgical side. These multidisciplinary pre operative briefing meetings have been introduced in limited specialties and should be rolled out throughout the. 3.6 Chief Executive/ Medical Director/ Director of Nursing tours The Chief Executive, Medical Director & Director of Nursing have instituted regular rounds to clinical areas, observing the quality of care, staff morale and listening to staff difficulties in delivering optimum clinical care. Clinical Director s and General Manager s are being encouraged to regularly tour their own areas. 4. What could be done? 4.1 Staffing review There is substantial evidence linking staffing levels to patient outcomes. a) Nursing staff: Two research papers, Nurse Staffing and Quality of Patient Care 5 and Outcomes of variation in hospital nurse staffing in English hospitals 6 have concluded that increased nurse staffing in hospitals was associated with lower hospital mortality and improved patient safety. 5 Kane RL, Shamliyan T, Mueller C, Duval S, Wilt T. Nursing Staffing and Quality of Patient Care. Evidence Report/Technology Assessment No. 151 (Prepared by the Minnesota Evidence based Practice Center under Contract No. 290 02 0009.) AHRQ Publication No. 07 E005. Rockville, MD: Agency for Healthcare Research and Quality. March 2007. 6 Anne Marie Rafferty et al., Outcomes of variation in hospital nurse staffing in English hospitals: Cross sectional analysis of survey data and discharge records, International Journal of Nursing Studies (2006), doi: 10.1016/j.ijnurstu.2006.08.003 5
Figure 2, compares East and North Hertfordshire NHS with a peer group of hospitals, correlating HSMR with nursing establishment, using publicly available data from Dr Foster. Figure 2 Nurses per 100 beds & HSMR Source Dr Foster 2005/06 300 250 245.1 212.8 200 150 158.4 143.3 171.2 163.2 185.4 172.6 148.5 141.2 Nurses per 100 beds (Dr Foster 05/06) HSMR (Dr Foster 05/06) 100 50 0 Cambridge Guy's & University Thomas's NHS Hospitals NHS Foundation Foundation Royal W est Sussex Royal United Hospital Bath Norfolk & Norwich University Hospital East & North Hertfordshire NHS Bedford Hospital NHS Luton & Dunstable NHS Foundation Barnet & Chase Farm Hospitals NHS West Hertfordshire Hospitals NHS A nursing staff review has already been undertaken and is under consideration by the Executive Team. b) Medical Staff A study by Professor Sir Brian Jarman, Explaining differences in English hospital death rates using routinely collected data 7 concluded that the ratios of doctors to head of population served, both in hospital and in general practice, seem to be a critical determinate of standardised hospital mortality rate. The more doctors per bed or per head of population, the lower the mortality rates. Using the same data source for benchmarking against a peer group as the nursing staff, initial findings show that the is an outlier in the number of doctors per 100 beds, having the smallest number at 47.1 whilst having an average HSMR of 98. 7, Jarman B, Gault S, Alves B, Hider A, Dolan S, Cook A, et al. Explaining differences in English hospital death rates using routinely collected data. BMJ. 1999;318:1515 1520.. (5 June.). 6
Figure 3 Doctors per 100 beds & HSMR Source: Dr Foster 2005/06 120 100 101.4 80 60 79.3 61.4 56.1 71.6 65.8 63 61.6 56.1 Drs per 100 beds (Dr Foster 05/06) HSMR (Dr Foster 05/06) 47.1 40 20 0 Cambridge University Hospitals NHS Foundation Guy's & Thomas's NHS Foundation Royal West Sussex Royal United Hospital Bath Norfolk & Norwich University Hospital East & North Hertfordshire NHS Bedford Hospital NHS Luton & Dunstable NHS Foundation Barnet & Chase Farm Hospitals NHS West Hertfordshire Hospitals NHS It is suggested that medical staff numbers are reviewed. The newly acquired Dr Foster data could be used to concentrate recruitment in areas where there is both high patient risk and high workload. 4.2 Process of Care Clinical deterioration can happen at any point in a patient s illness, or care process, but patients are particularly vulnerable following an emergency admission to hospital, after surgery and during recovery from critical illness. Analysis of 576 deaths reported to the National Patient Safety Agency 8 (NPSA) national reporting and learning system over a 1 year period (2005), identified that 11% were as a result of deterioration not recognised or acted upon. The majority of these incidents can be attributed to three themes: Observations not made for a prolonged period and therefore changes in a patient s vital signs not detected No recognition of the importance of the deterioration and/or no action taken other than recording of observations Delay in the patient receiving medical attention, even when deterioration has been detected and recognised Building a reliable system to recognise the signs, to report and respond to them will have a significant impact in reducing mortality. This system needs to include: a) Reliable ward observations There has been awareness since the middle of the 1990 s that acute deterioration is apparent in two thirds of patients prior to cardiac arrest. 9 Simple observations of 8 National Patient Safety Agency. Recognising and responding appropriately to early signs of deterioration in hospitalised patients. November 2007 9 Franklin C. Mathew J. (1994) Developing strategies to prevent in hospital cardiac arrest: analysing responses of physicians and nurses in the hours before the event. Crit Care Med,22:244 7 7
respiratory rate, heart rate, systolic blood pressure and level of consciousness will help to identify at risk patients. Evidence suggests that respiratory observations are often omitted, and that the respiratory rate has lost emphasis in training, possibly due to technological developments such as dinamaps and pulse oximeters. Training and refreshment of observation skills, especially respiratory rates, should be cascaded throughout the. b) A trigger mechanism Monitoring patient observations is not sufficient and there must be a robust mechanism for recognising and responding to changes. The already uses Early Warning Scores but these should be reviewed and simplified. Luton and Dunstable NHS has developed a colour banded card that removes the need for the score to be calculated. The information is explicit if the patient s condition is in decline, scores fall into red or yellow boxes. If two yellow, or one red box, are filled, the nurse must take action and respond to the patient s condition. Nurses can also call for assistance if they think the patient is showing signs of deterioration, even though they have not triggered a yellow or red score on the chart. The use and appropriateness of the existing Early Warning Scores system is currently under review by the Critical Care Delivery Group. c) An assertive reporting/communication tool SBAR The introduction of a reporting or communication tool will help staff to deliver information effectively to busy clinical staff who may not appreciate the importance of the call. A tool called Situation Background Assessment Recommendation (SBAR), first used by the US Navy and subsequently developed by Kaiser Permanente and the IHI in the US, is an easy to remember mechanism that is used to frame conversations, especially critical ones, requiring a clinician s immediate attention and action. It acts as a reminder to the reporting clinician to describe the situation, background, assessment and recommendation and give essential information. It allows the clinician receiving the information to know the order and context of that information and gives clarity to what action is required of him/her. It also gives the nurses confidence in their judgement to call for medical assistance and helps them to be more assertive in requesting an immediate response when needed. d) Critical Care Outreach Team Once deterioration is recognised a rapid and effective response is required. An efficient way of providing an appropriate multidisciplinary response to the deteriorating patient is the use of a Critical Care outreach team. A Critical Care outreach service is a multi disciplinary approach to the identification and management of patients at risk of, or developing critical illness. One of the service s main roles is education and facilitation of ongoing learning for staff to care for critically ill patients. National guidance recommends that hospitals develop an outreach service to support the wards in managing patients at risk. 10, 11 Implementing an outreach service will enable the to adhere to the various guidelines and recommendations produced nationally, ensuring that timely and appropriate care is given to the critically ill patient. Cardiac arrest is frequently a fatal event. There is good evidence that rapid response by appropriate clinicians prevents cardiac arrest. Figure 4 shows the impact that introducing a critical care outreach team responding in a timely manner to effective 10 The National Confidential Enquiry into Patient Outcome and Death. 2005, NCEPOD 11 An acute problem recognition of and response to acute illness in adults in hospital.2007. National Institute for Health and Clinical Excellence (NICE) 8
early warning systems had on reducing the incidence of cardiac arrest calls at the Middlesex Hospital. Figure 4 Incidence of Inpatient cardiac arrest calls 2000 2004 (Middlesex Hospital) 120 100 80 60 40 20 0 M/sex 2000 M/sex 2001 M/sex 2002 M/Sex 2003 M/Sex 2004 Ref: Reducing avoidable deaths in hospital, Sheila Adam, April 2008 A Case of Need for a Critical Care outreach service was submitted to the Board in April 2008 and a phased approach to implementing this service has been approved. e) Increase Critical Care capacity Current demand for Critical Care beds is high. The occupancy rates for Intensive Care across both of the hospital sites has risen over the past three years with the average rate rising from 85% in 2006/7 to 95% in 2007/8. This rise in occupancy, coupled with longer lengths of stay may result in poorer outcomes. There is good evidence that the ready availability of HDU and ICU beds enables sick or deteriorating patients to be admitted early to these units. This leads to better outcomes and saves lives. Copeland has shown that the mortality rate for a patient where there is an HDU bed available only 50% of the time is 20% higher than when there is an HDU bed immediately available 90% of the time (see fig. 5). Other investigators have shown that mortality is higher on general wards than specialist wards. Figure 5 illustrates the decrease in mortality outcome over expected ratios with increased critical care capacity. 9
Figure 5 HDU effects HDU availability Mortality O/E Ratio Morbidity O/E Ratio 100% 0.97 0.98 90% 0.99 0.99 70% 1.08 1.06 50% 1.20 1.08 Ref: Graham Copeland, Using clinical audit to understand and improve surgical deaths, 9 th April 2008 * O/E is observed over expected A business case proposing expansion of critical care facilities by implementing six surgical step down beds is in the process of being developed. The Board have recently approved the expansion of critical care facilities and 6 surgical step down beds are to be developed on Level 7 on the Lister site. 4.3 Role out mortality monitoring by directorate Up until last year the did not routinely monitor mortality rates. Last year CHKS was engaged to supply risk adjusted mortality rates for all clinical areas. These have subsequently been monitored and areas of concern investigated. The Tust has also now also engaged Dr Foster to supply it with a wide variety of clinical information which includes both risk adjusted mortality and HSMR. In order to make best use of this data a systematic approach to monitoring and acting on this data needs to be introduced. The Board has already determined that it wishes to review crude mortality and HSMR on a regular basis and that the Medical Director would be tasked with overseeing the 's process for monitoring mortality. A preliminary meeting with Dr Foster and clinical representatives has already taken place. It is envisaged that a regular Clinical Quality Report will be produced from the Dr Foster and CHKS data and will be forwarded to each directorate for review and action. This Clinical Quality Report will then be reviewed at the directorate's Performance Management meetings. This will enable a multidisciplinary approach to clinical quality, drawing together representatives from general, medical and nursing management and will allow performance management of the directorate using the Clinical Quality Report as a benchmark. The timing of the introduction of this process will depend on the availability of adequate data analyst support but it is hoped to start within the next three months. 4.4 Leadership safety walk around The Chief Executive and Medical Director have already carried out tours of the Paediatric, Obstetric, General Medicine, Care of the Elderly and Accident and Emergency departments and will continue to tour clinical areas. The Director of Nursing also has regular clinical leadership walk arounds. Senior Managers need to be demonstrating to their staff that saving lives and patient safety is a key priority for the and, as such, is important to them. The s vision is that it is committed to being trusted by the community it serves to provide a 10
comprehensive range of modern, convenient high quality and caring acute hospital services. Visible presence of the Chief Executive, Clinical Directors and General Managers leading walk arounds specifically addressing patient safety issues will communicate the message and help to bring about a cultural change in raising the profile of saving lives. 12 It is recommended that the programme of clinical walk arounds are systematically reviewed and that these become routine for all clinical managers within the. 5. Audit, structure & support The s patient safety structure will need development and support to: Enhance our Patient Safety culture Support an increase in requirements for audit & information of patient safety indicators within the Identify Clinical Champions for initiatives Embed clinical performance in the performance management process Address multi disciplinary training needs. Health organisations with low mortality have well developed systems for monitoring and performance managing patient safety. As a minimum this should consider regular monitoring and performance management of the following key indicators: i. HSMR ii. Crude mortality by clinical division iii. Risk adjusted mortality by specialty and by clinical team iv. Compliance with patient observations of vital signs v. Response time for deteriorating patients by specialty team vi. Cardiac arrest rate vii. Chart review of all cardiac arrests by specialty Setting up these processes will require initial help from a statistician, data information analyst and audit staff. Ongoing performance management will be overseen by the Director of Nursing and Medical Director and delegated to Clinical Director s, Matrons and General Manager s. They will need adequate support from audit staff and information analysts. Luton & Dunstable NHS, a national leader in improving patient safety, has a team comprising the Chief Executive, the Medical Director, and two Patient Safety Project Managers leading the improvements. 6. Key Goals Patient safety is a key priority for the and as such a transformational, proactive programme will help to achieve a culture of patient safety, improve patient care and reduce mortality.. It is generally accepted that approximately 1 in 10 deaths in first world hospitals could be prevented by better quality medical care. 13 The NHS Institute for Innovation and Improvement has found that s working on reducing mortality have achieved reductions in Hospital Standardised Mortality Rate (HSMR) of up to 20%, even when their starting HSMR was below average. They state that a reduction of just 10% across England would mean 10,000 lives saved per year. 12 Stephen Ramsden. Prioritising the reduction of avoidable mortality: initiating and leading change Healthcare Event April 2008 13 Jarman, B, et al, Explaining differences in English hospital death rates using routinely collected data BMJ, 318:1515 1520, 1999 11
It is recommended that the Board consider setting mortality reduction targets as a means to monitoring performance and driving forward change. Suggested goals include: Reduce the s crude mortality rate to below 2.0%. Reduce HSMR to 95 by June 2009, 90 by June 2010 (based on 06/07 figure) The currently has average mortality but it could and should aspire to have a mortality rate substantially below average. To achieve this will require a variety of initiatives and some investment as described within this paper. There is substantial scientific evidence available for what works in reducing mortality and extra investment in this area is highly likely to bring a return in terms of extra lives saved. 12