- Email: [email protected]
Outcome from intensive care and measuring performance
CRITICAL ILLNESS AND INTENSIVE CARE: I
Outcome from intensive care and measuring performance
To acquire a detailed understanding of the critical care pathway we must analyse each individual step along the way. Measurement of performance, or quality, at each stage can identify key steps, or quality indicators that have an influence on the eventual patient outcome. Defined by the Intensive Care Society (ICS) as ‘.standards, which if upheld will likely improve the quality of patient care by means of improved safety, better patient outcomes or greater efficiency’, quality indicators should be: evidence based relatively easy to collect used to drive improvement and therefore measured and studied at frequent time intervals sophisticated enough to adjust for varying levels of risk in the population considered.6 Intensive care quality indicators have been endorsed by many national bodies including the Department of Health,1 ICS,6 The Joint Commission,7 and The Australian Council of Healthcare Standards8 to name but a few (Table 1). A significant challenge in developing quality indicators is to strike a balance between the number, validity and reliability of measures and the burden of data collection. The introduction of clinical information systems with automated data collection will clearly reduce this burden. Quality indicators should not be confused with ‘core standards.’ Core standards are national or regional minimum standards for critical care provision and can vary between country and region. There is huge overlap between these two concepts and indeed one departmental quality indicator could be compliance to these core standards. In the UK ‘Core Standards for Intensive Care Units’ was published in 2013 by the Faculty of Intensive Care Medicine and represents current UK minimum standards for critical care provision. Whether dealing with minimum standards or quality indicators we must acknowledge the continuous development and evolution of the measured indices. As critical care progresses, treatments change and political priorities shift and there is a continuous requirement to re-evaluate the indices you are measuring. For example, the UK core standards document is already planned for replacement by a ‘Good Practice in Intensive Care Services’ publication which will be reviewed and updated on an annual basis.
Rob Whittle Jonathan Shelton
Abstract Critical care outcomes are becoming increasingly important in the modern NHS and frequently ICU performance is judged in this manner. However the eventual outcome of any individual patient or patient group is dependent upon a complex process that precedes the end-point. This article reviews the aspects of ICU structure, process and outcome that can be used as quality indicators or to measure performance.
Keywords Critical care; intensive care; outcome; performance; quality
Introduction As doctors working in the modern NHS we all appreciate the importance of healthcare outcomes, be that mortality, morbidity or alternative measures; and the performance of many is now judged in this manner by patients, relatives and healthcare purchasers. However, the eventual outcome of any individual patient or patient group is dependent upon an extremely complex process preceding the end-point and failure at any stage can influence the result. Therefore an in-depth understanding of performance throughout the entire patient pathway is necessary to influence the outcome.
Performance [ Quality Good performance has been emphasized by the Department of Health in the recent publications ‘Quality Critical Care’,1 ‘High Quality Care For All’2 and ‘Equity and Excellence: Liberating the NHS’;3 and, although not identical, the term ‘measuring quality’ has become synonymous with ‘measuring performance’. Quality in healthcare has been defined by The Institute of Medicine as ‘the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge’.4
Measuring quality and quality indicators
Quality indicators e structure
Measurement of performance, or quality, has been divided by Donabedian into three classic components: structure, process and outcome5; and in latter years this construct has been widely accepted by the medical profession. Structure represents the way we organize care, process refers to what we do for patients and their families and outcome denotes the final results we achieve.
The multidisciplinary team The modern ICU multidisciplinary team (MDT) led by a consultant intensivist may include several intensive care trainees, acute critical care practitioners, senior and junior nurses, specialist ICU physiotherapists, pharmacists, dieticians, speech and language therapists and clerical staff. Conduct of daily ward rounds involving the MDT has been shown to reduce the odds of death in a large population-based retrospective cohort study9 and pharmacist participation is associated with fewer adverse drug events.10
Rob Whittle FRCA EDIC is a Specialist Registrar in Anaesthesia and Intensive Care in the Northern Deanery, UK. Conflicts of interest: none declared. Jonathan Shelton FRCA DICM FCICM FFICM is a Consultant in Anaesthesia and Intensive Care at Royal Victoria Infirmary, Newcastle upon Tyne, UK. Conflicts of interest: none declared.
SURGERY 33:4
Closed units Intensive care units can be classified as open or closed. In an open unit any physician is able to admit and care for patients in
180
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
Nurse:patient ratio Recommended nurse:patient ratios are dependent upon patient severity of illness or level of care12,13 (Table 2). There is increasing evidence that the higher the ratio the better the quality of care and the fewer the number of unit-acquired infections.14
Endorsed quality indicators Quality indicator STRUCTURE
C C C C C
C
C C C C C C C
PROCESS
C C C C C C
OUTCOME
C C
C C C C C C C C
C
C
Multidisciplinary ward rounds Intensivist cover 24/7 Nurse:patient ratio Daily review and management plan Participation in national comparative audit Outreach and track and trigger systems Rehabilitation Patient isolation Bed availability/occupancy Daily goals sheet Structured handover Quality Improvement Programme Fulfils national provision for ICM Hand hygiene Appropriate early enteral nutrition End of life care pathway Care bundles Stress ulcer prophylaxis DVT prophylaxis Standardized mortality ratio Review of morbidity and mortality/ adverse event reporting ICU length of stay Readmissions within 48 hours Discharges from 22:00e07:00 Non-clinical transfers Unit-acquired bacteraemias Unit-acquired MRSA Unit-acquired C. difficile or MRSA Catheter-related bloodstream infection Rate of ventilator-associated pneumonia Unplanned extubations/reintubation rate
National body
C
I, D, E I, A, E I S S, A, D
C
D
C
D I I, S, E I I, E D, E I, E I I I, S D, S, J J J I, S, D I, S, E
C C C C
C C C C C C C C C C C C C C
C
J I, I, I I, I, I, I,
C
E
C
E
C C C C C C C
Intensive care outreach and track and trigger systems The concept of critical care without walls was introduced in 2000 in the Department of Health publication ‘Comprehensive Critical Care’,13 and refers to a comprehensive hospital-wide approach with services that extend beyond the physical boundaries of intensive care. Introduction of critical care outreach and track and trigger systems in a stepped wedge randomized controlled trial has been associated with a significant reduction in mortality15 but evidence is inconsistent and a Cochrane review highlights the poor methodological quality of studies and the need for further trials. Bed availability Units running at full occupancy will not always be able to accept referred patients, which will increase the number of non-clinical inter-hospital transfers, and in an observational study was shown to be associated with increased mortality. Communication The conduct of a structured handover between outgoing and incoming medical teams and the use of a patient daily goals sheet have been shown to improve communication, reduce adverse events and reduce ICU length of stay.
A S
Levels of care and nurse:patient ratio
J, S S E J, E
Level of care
Description
Nurse:patient ratio
0
Patients whose needs can be met through normal ward care in an acute hospital Patients at risk of their condition deteriorating, or those recently relocated from higher levels of care, whose needs can be met on an acute ward with additional advice and support from the critical care team Patients requiring more detailed observation or intervention including support for a single failing organ system or postoperative care and those ‘stepping down’ from higher levels of care Patients requiring advanced respiratory support alone or basic respiratory support together with support of at least two organ systems. This level includes all complex patients requiring support for multi-organ failure
1:4e1:12
1
I ¼ Intensive Care Society. D ¼ Department of Health. S ¼ Scottish Intensive Care Society. J ¼ Joint Commission. A ¼ Australian Council of Healthcare Standards. E ¼ European Society of intensive Care Medicine.
2
Table 1
the ICU. The physician makes all decisions during the patient’s stay and an intensivist may or may not be consulted. Closed units operate under the control of an intensivist who makes decisions about admission, discharge and daily goals after consultation with the specialist team. Systematic review has confirmed that patients managed in closed units have lower mortality, fewer complications and shorter ICU and hospital stays.11 The vast majority of ICUs in the UK operate as closed units.
SURGERY 33:4
3
1:4e1:12
1:2
1:1
Table 2
181
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
National audit Input of data into national audit programmes is the cornerstone of any quality improvement programme.
unit’s mortality rate is vital if mortality is to be used as a valid outcome measure. Standardized mortality ratio (SMR) is the term applied to this ‘corrected’ measure of mortality:
Standardized Mortality Ratio ¼ Actual Mortality Rate=Expected Mortality Rate
Quality indicators e process
SMR allows comparisons of mortality figures between critical care units and within a single unit over a period of time. For example, if an SMR for a particular unit is more than 1, this would suggest that more people than expected were dying within that unit for its case mix and disease severity at presentation. This measure of worse outcomes for patients within a unit when compared with the expected mortality rate, potentially allows for the identification of areas of care in need of improvement. These could then be addressed and SMR re-assessed to confirm an actual clinical benefit to patients. Consequently, it becomes possible to monitor an individual department’s performance at any time and, in addition, to track its outcomes over time. SMR relies on an accurate prediction of mortality for a clinical presentation. For this we use critical care scoring systems.
Care bundles A care bundle is defined by the Institute for Healthcare Improvement as ‘a small straightforward set of evidencer-based practices, generally three to five, that when performed collectively and reliably, have been proven to improve patient outcomes’. Several care bundles are in common use including the ventilator care bundle, the central venous line bundle and the sepsis care bundle. The ventilator care bundle has four aspects: raising the head of the patients bed to 30e40 stress ulcer prophylaxis DVT prophylaxis spontaneous breathing trials. Use of this bundle has been shown to reduce the incidence of ventilator-associated pneumonia. The central venous line bundle is a set of five simple steps that have been proven to reduce the incidence of catheter-related bloodstream infection. The sepsis care bundle developed by the Surviving Sepsis Campaign is divided into early and late goals and has been associated with a reduction in mortality in a small cohort study.
Scoring systems Scoring systems are applied throughout clinical practice and critical care. They can be generic (applied to all disease presentations) or specific (applied to a single disease process). They can be anatomical (e.g. radiological grading of sub-arachnoid haemorrhage) or physiological (e.g. APACHE II). They can look at disease severity on admission to critical care or track the patient at regular intervals throughout the clinical course. An ideal critical care scoring system should be quick and easy to perform, involve minimal recording bias and have good discrimination and calibration. Discrimination is the ability of the scoring system to predict mortality in individuals without ambiguity. Currently all generic critical care scoring systems are poor discriminators and are unreliable when applied to individual patients when trying to predict an individual’s mortality. Calibration is the similarity between predicted outcome and actual outcome within a population. It is calibration that is important when using scoring systems to calculate SMRs and currently there are several critical care scoring systems that show an acceptable level of calibration and can be used to calculate SMR. The scoring systems used to calculate expected mortality rates and therefore SMR within a population are generic and are developed by prospectively collecting data sets from population groups where mortality is recorded. They apply a relative weighting to each patient variable, measure physiological derangement and take into account diagnosis and co-morbidities. Unfortunately, the validity of these scoring systems is degraded when they are developed on a population group other than the one to which it is being applied and over time as critical care practices and therapy change and develop. One of the most widely used scoring systems worldwide is the Acute Physiology and Chronic Health Evaluation score (APACHE II). This was developed by Knaus et al. and published in 1985. It used data from 5815 critical care admissions in the USA between 1979 and 1982. Despite widespread use its application to current
Early enteral nutrition The use of early enteral nutrition is associated with improved nutritional intake and trends towards reduced infectious complications and mortality. Several national practice guidelines recommend its appropriate use. End of life care pathways Both the Department of Health and the Robert Wood Foundation in the USA have defined quality markers for end of life care: patient- and family-centred decision-making communication continuity of care emotional and practical support symptom management and comfort care spiritual support emotional and organizational support for intensive care clinicians. In the UK the Liverpool Care Pathway is the most common tool incorporating these domains. Quality indicators e outcome Assessing and measuring outcome is a key aspect of evaluating the quality of critical care delivery. Mortality can be used as an easily definable measure of performance whereas morbidity measures can be more complex to define and interpret. Mortality Crude mortality measures are generally accepted as a poor measure of quality since they do not take into account variables such as case-mix, patient co-morbidities and disease severity. Correcting for these variables and therefore ‘standardizing’ a
SURGERY 33:4
182
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
receiving step down patients have minimal staff. Studies have shown they are associated with increased morbidity and mortality.
critical care patients in the UK could be questioned. Firstly, it was developed by analysing patient data from North America and secondly, these data are now over 30 years old. These criticisms, unfortunately, are limitations of many of the scoring systems within critical care.
Readmission rates Intensive care readmission rates range from 4 to 14%. Compared to patients with a similar severity of illness readmitted patients have twice the ICU length of stay and a two- to tenfold increased risk of death.
The Intensive Care National Audit and Research Centre (ICNARC) ICNARC was set up in 1994 alongside the ICS. Its goal was to ‘foster improvements in the organisation and practice of critical care within the UK.’ Currently more than 90% of adult general critical care units participate. The ICNARC model originally used data from over 200,000 admissions to the national case mix programme database and, with similarities to other scoring systems like APACHE, is now able to produce an expected mortality rate for a population. It combines a physiology score during the first 24 hours of critical care admission with patient demographics, reason for admission, co-morbid diagnoses, source of admission, surgical urgency and whether or not cardiopulmonary resuscitation (CPR) has taken place in the preceding 24 hours prior to admission. Its use for the UK population is valid since it has been generated using data from that population and it is calibrated regularly. The ICNARC case mix programme delivers each participant unit quarterly electronic data analyses reports. This report includes the SMR for the unit in addition to a myriad of morbidity data.
Healthcare-acquired infection Reducing the incidence of hospital-acquired infections has been a priority for the NHS as set out in the 2008/09 NHS Operating framework and the 2007 Public Service Agreement ‘Ensure better care for all.’ Mandatory surveillance has been in place since 2004, with a target reduction of 30% by 2010/11 compared with 2007/08 figures. Patients who have been in units for 48 hours or more and who screen positive are considered to have unitacquired infection. Long-term morbidity quality indicators In recent years there has been significant progress made in this previously under-explored area of long-term critical care morbidity outcomes. One-year survival following critical care in the UK stands at around 60%. It has been shown that amongst these survivors they continue to not only have a persistently raised mortality rate when compared with a matched population, but also suffer from increased rates of physical and psychological disease. The estimated prevalence of disease following critical care is: physical half of all critical care survivors have some degree of dependency in activities of daily living at 6 months 30% of survivors of working age will not return to work psychological anxiety 12e43% depression 10e30% symptoms of post-traumatic stress disorder 5e64%. However, quality trials in this area are still lacking and it remains unclear as to the exact prevalence of disease in this group of patients. Efforts are being made currently via the Intensive Care Outcome Network study (ICON) and the Intensive Care After Network (ICANUK) to recruit patients to try to answer this question in more detail.
Morbidity It is generally accepted that SMR is the key clinical outcome quality indicator for adult critical care. In recent years however, other outcome indicators have been explored as potential markers of quality. Short-term morbidity quality indicators Examples of these outcome measures would include: length of critical care admission ventilator days night-time discharges readmission rates non-clinical transfers healthcare-acquired infection rates incidence of unit-acquired meticillin-resistant Staphylococcus aureus (MRSA) bacteraemias incidence of all unit-acquired bacteraemias unit-acquired MRSA or Clostridium difficile incidence of ventilator-acquired pneumonia. Many of these markers are the source of public health, national or local targets (e.g. hospital-acquired infection rates); some are useful in order to collect data of incidence and prevalence of clinical conditions nationally (e.g. admission following CPR). Much of the data is collected and reported on behalf of UK critical care by ICNARC. However, data validating the use of many of these measures as a marker of quality are lacking. Currently, there is an assumption that these surrogate end-points potentially could be associated with long-term benefits.
Intensive care follow-up clinics Critical care follow-up clinics are becoming increasingly available in many hospitals. They provide opportunity for local audit of practice but also the opportunity of larger scale data collection that may facilitate future studies like ICON. Exactly what should be assessed and how is still a source of debate but standardization is beginning to occur. Longer-term outcome measures that may be assessed following critical care discharge include: physiological respiratory volumes (e.g. vital capacity) gas flow (e.g. FEV1) pulmonary diffusion capacity upper airway visualization (e.g. bronchoscopy)
Night time discharges Discharges from intensive care units during night shift hours are not planned, handover is frequently less than optimal and wards
SURGERY 33:4
183
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
physical functional status assessment of activities of daily living (e.g. Katz’s measurement) mental and neuropsychological functional status anxiety and depression (e.g. HADS survey) post-traumatic stress disorder (e.g. PCL-C questionnaire) health-related quality of life measures Short Form Health Survey (SF-36), Sickness Impact Profile scale (SIP) Perceived Quality of Life scale (PQOL) (38).
terms of length and quality of life. The cost of particular treatments can then be compared in quality adjusted life years (QALYs). A recent review of the cost-utility of intensive care treatment estimated it to be approximately £7000 per QALY which compares favourably with other acute care interventions.
Summary Crude and standardized mortality measures were the traditional markers of whether or not a critical care was achieving a satisfactory outcome for its patients. However, this has been increasingly shown not to be the whole story and markers of morbidity both in the short and long term as well as following up patients after critical care is increasingly being seen as a valid approach to the assessment of quality of care. If we are to improve outcomes for critical care patients we must improve quality in all three areas of structure, process and outcome using an organized Quality Improvement Programme. We must also be vigilant and adapt as markers of quality will change, as our speciality develops into the future. A
Satisfaction surveys Satisfaction surveys have long been seen as a potential measure of quality in healthcare services. Critical care provides additional challenges to this form of assessment since many patients will have altered consciousness and poor recollection of their critical care stay. In addition, a patient’s family often play a more prominent role in a critical care admission. Critical care physicians may frequently spend longer periods of time discussing clinical decisions and care plans with a patient’s relatives than with the patient themselves. Assessing a patient and their family’s satisfaction with the critical care service they received is felt to be of increasing importance when assessing overall quality of care. This satisfaction has been characterized into six general domains: assurance: the need to feel hope for a desired outcome information: the need for consistent, regular and realistic updates proximity: the need to be physically and emotionally near the patient support: the need for support services if required comfort: the need for comfortable surroundings and waiting areas benefit: the need to feel that the treatments are in the patient’s best interest. Many of these satisfaction domains have been validated when assessed via post-care questionnaire and are now a part of routine critical care follow-up.
REFERENCES 1 Department of Health. Quality critical care e beyond ‘comprehensive critical care’. London: Department of Health Publications, 2005. 2 Department of Health. High quality care for all: NHS next stage review final report. London: Department of Health Publications, 2008. 3 Department of Health. Equity and excellence: liberating the NHS. London: Department of Health Publications, 2010. 4 Committee on Quality of Health Care in America. Crossing the quality chasm: a new health system for the 21st Century. Washington, DC: National Academy Press, 2001. 5 Donabedian A. Evaluating the quality of medical care. Milbank Q 2005; 83: 691e729. 6 Intensive Care Society. ICS quality indicators 2010. Available at: http://www.ics.ac.uk/intensive_care.../standards.../ics_quality_ indicators. 7 The Joint Commission. National hospital quality measures e ICU. http://www.jointcommission.org/PerformanceMeasurement/ MeasureReserveLibrary/SpecþManualþ-þICU.htm. 8 Australian Council on Healthcare Standards. Australasian clinical indicator report 1998e2006. 2007. Ultimo, Australia: ACHC, http:// www.achs.org.au/pdf/Australasian_CIR_8th_Edition.pdf. 9 Kim MM, Barnato AE, Angus DC, Fleisher LF, Kahn JM. The effect of multidisciplinary care teams on intensive care unit mortality. Arch Intern Med 2010; 170: 369e76. 10 Kane SL, Weber RJ, Dasta JF. The impact of critical care pharmacists on enhancing patient outcomes. Intensive Care Med 2003; 29: 691e8. 11 Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. J Am Med Assoc 2002; 288: 2151e62. 12 BACCN, CC3N, RCN. Standards for nurse staffing in critical care. September 2009, http://www.baccn.org.uk/downloads/BACCN_ staffing_document.pdf. 13 Department of Health. Comprehensive critical care. London: Department of Health Publications, 2000. 14 Kane RL, Shamliyan T, Mueller C, Duval S, Wilt T. Nursing staffing and quality of patient care. Evidence Report/Technology assessment No.
Audit and quality improvement programmes In 2002 the National Institute for Clinical Improvement defined clinical audit as ‘a quality improvement process that seeks to improve patient care and outcomes through systematic review of care against explicit criteria and the implementation of change’. The audit cycle involves identification of the problem, defining a standard, data collection, comparing performance against that standard and implementing change. The cycle is then repeated to assess the response to change. Application of the audit cycle to the aforementioned quality indicators can be used to improve patient outcome and is known as a Quality Improvement Programme.
Economic performance Measuring clinical performance cannot be completely divorced from measuring economic performance, which is achieved by the use of cost-utility analysis. This specific form of cost-effectiveness analysis measures cost in the form of monetary units but benefit in
SURGERY 33:4
184
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
15 Rhodes A, Moreno RP, Azoulay E, et al. Prospectively defined indicators to improve the safety and quality of care for critically ill patients: a report from the Task Force on Safety and Quality of the European Society of Intensive Care Medicine (ESICM). Intensive Care Med 2012; 38: 598e605.
151 (Prepared by the Minnesota evidence based practice Center under Contract No. 290-02-0009.). March 2007. Rockville, MD: Agency for Healthcare Research and Quality. AHRQ Publication No. 07-E005, http://www.ahrq.gov/downloads/pub/evidence/pdf/ nursestaff/nursestaff.pdf.
SURGERY 33:4
185
Ó 2015 Elsevier Ltd. All rights reserved.
Outcome from intensive care and measuring performance
To acquire a detailed understanding of the critical care pathway we must analyse each individual step along the way. Measurement of performance, or quality, at each stage can identify key steps, or quality indicators that have an influence on the eventual patient outcome. Defined by the Intensive Care Society (ICS) as ‘.standards, which if upheld will likely improve the quality of patient care by means of improved safety, better patient outcomes or greater efficiency’, quality indicators should be: evidence based relatively easy to collect used to drive improvement and therefore measured and studied at frequent time intervals sophisticated enough to adjust for varying levels of risk in the population considered.6 Intensive care quality indicators have been endorsed by many national bodies including the Department of Health,1 ICS,6 The Joint Commission,7 and The Australian Council of Healthcare Standards8 to name but a few (Table 1). A significant challenge in developing quality indicators is to strike a balance between the number, validity and reliability of measures and the burden of data collection. The introduction of clinical information systems with automated data collection will clearly reduce this burden. Quality indicators should not be confused with ‘core standards.’ Core standards are national or regional minimum standards for critical care provision and can vary between country and region. There is huge overlap between these two concepts and indeed one departmental quality indicator could be compliance to these core standards. In the UK ‘Core Standards for Intensive Care Units’ was published in 2013 by the Faculty of Intensive Care Medicine and represents current UK minimum standards for critical care provision. Whether dealing with minimum standards or quality indicators we must acknowledge the continuous development and evolution of the measured indices. As critical care progresses, treatments change and political priorities shift and there is a continuous requirement to re-evaluate the indices you are measuring. For example, the UK core standards document is already planned for replacement by a ‘Good Practice in Intensive Care Services’ publication which will be reviewed and updated on an annual basis.
Rob Whittle Jonathan Shelton
Abstract Critical care outcomes are becoming increasingly important in the modern NHS and frequently ICU performance is judged in this manner. However the eventual outcome of any individual patient or patient group is dependent upon a complex process that precedes the end-point. This article reviews the aspects of ICU structure, process and outcome that can be used as quality indicators or to measure performance.
Keywords Critical care; intensive care; outcome; performance; quality
Introduction As doctors working in the modern NHS we all appreciate the importance of healthcare outcomes, be that mortality, morbidity or alternative measures; and the performance of many is now judged in this manner by patients, relatives and healthcare purchasers. However, the eventual outcome of any individual patient or patient group is dependent upon an extremely complex process preceding the end-point and failure at any stage can influence the result. Therefore an in-depth understanding of performance throughout the entire patient pathway is necessary to influence the outcome.
Performance [ Quality Good performance has been emphasized by the Department of Health in the recent publications ‘Quality Critical Care’,1 ‘High Quality Care For All’2 and ‘Equity and Excellence: Liberating the NHS’;3 and, although not identical, the term ‘measuring quality’ has become synonymous with ‘measuring performance’. Quality in healthcare has been defined by The Institute of Medicine as ‘the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge’.4
Measuring quality and quality indicators
Quality indicators e structure
Measurement of performance, or quality, has been divided by Donabedian into three classic components: structure, process and outcome5; and in latter years this construct has been widely accepted by the medical profession. Structure represents the way we organize care, process refers to what we do for patients and their families and outcome denotes the final results we achieve.
The multidisciplinary team The modern ICU multidisciplinary team (MDT) led by a consultant intensivist may include several intensive care trainees, acute critical care practitioners, senior and junior nurses, specialist ICU physiotherapists, pharmacists, dieticians, speech and language therapists and clerical staff. Conduct of daily ward rounds involving the MDT has been shown to reduce the odds of death in a large population-based retrospective cohort study9 and pharmacist participation is associated with fewer adverse drug events.10
Rob Whittle FRCA EDIC is a Specialist Registrar in Anaesthesia and Intensive Care in the Northern Deanery, UK. Conflicts of interest: none declared. Jonathan Shelton FRCA DICM FCICM FFICM is a Consultant in Anaesthesia and Intensive Care at Royal Victoria Infirmary, Newcastle upon Tyne, UK. Conflicts of interest: none declared.
SURGERY 33:4
Closed units Intensive care units can be classified as open or closed. In an open unit any physician is able to admit and care for patients in
180
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
Nurse:patient ratio Recommended nurse:patient ratios are dependent upon patient severity of illness or level of care12,13 (Table 2). There is increasing evidence that the higher the ratio the better the quality of care and the fewer the number of unit-acquired infections.14
Endorsed quality indicators Quality indicator STRUCTURE
C C C C C
C
C C C C C C C
PROCESS
C C C C C C
OUTCOME
C C
C C C C C C C C
C
C
Multidisciplinary ward rounds Intensivist cover 24/7 Nurse:patient ratio Daily review and management plan Participation in national comparative audit Outreach and track and trigger systems Rehabilitation Patient isolation Bed availability/occupancy Daily goals sheet Structured handover Quality Improvement Programme Fulfils national provision for ICM Hand hygiene Appropriate early enteral nutrition End of life care pathway Care bundles Stress ulcer prophylaxis DVT prophylaxis Standardized mortality ratio Review of morbidity and mortality/ adverse event reporting ICU length of stay Readmissions within 48 hours Discharges from 22:00e07:00 Non-clinical transfers Unit-acquired bacteraemias Unit-acquired MRSA Unit-acquired C. difficile or MRSA Catheter-related bloodstream infection Rate of ventilator-associated pneumonia Unplanned extubations/reintubation rate
National body
C
I, D, E I, A, E I S S, A, D
C
D
C
D I I, S, E I I, E D, E I, E I I I, S D, S, J J J I, S, D I, S, E
C C C C
C C C C C C C C C C C C C C
C
J I, I, I I, I, I, I,
C
E
C
E
C C C C C C C
Intensive care outreach and track and trigger systems The concept of critical care without walls was introduced in 2000 in the Department of Health publication ‘Comprehensive Critical Care’,13 and refers to a comprehensive hospital-wide approach with services that extend beyond the physical boundaries of intensive care. Introduction of critical care outreach and track and trigger systems in a stepped wedge randomized controlled trial has been associated with a significant reduction in mortality15 but evidence is inconsistent and a Cochrane review highlights the poor methodological quality of studies and the need for further trials. Bed availability Units running at full occupancy will not always be able to accept referred patients, which will increase the number of non-clinical inter-hospital transfers, and in an observational study was shown to be associated with increased mortality. Communication The conduct of a structured handover between outgoing and incoming medical teams and the use of a patient daily goals sheet have been shown to improve communication, reduce adverse events and reduce ICU length of stay.
A S
Levels of care and nurse:patient ratio
J, S S E J, E
Level of care
Description
Nurse:patient ratio
0
Patients whose needs can be met through normal ward care in an acute hospital Patients at risk of their condition deteriorating, or those recently relocated from higher levels of care, whose needs can be met on an acute ward with additional advice and support from the critical care team Patients requiring more detailed observation or intervention including support for a single failing organ system or postoperative care and those ‘stepping down’ from higher levels of care Patients requiring advanced respiratory support alone or basic respiratory support together with support of at least two organ systems. This level includes all complex patients requiring support for multi-organ failure
1:4e1:12
1
I ¼ Intensive Care Society. D ¼ Department of Health. S ¼ Scottish Intensive Care Society. J ¼ Joint Commission. A ¼ Australian Council of Healthcare Standards. E ¼ European Society of intensive Care Medicine.
2
Table 1
the ICU. The physician makes all decisions during the patient’s stay and an intensivist may or may not be consulted. Closed units operate under the control of an intensivist who makes decisions about admission, discharge and daily goals after consultation with the specialist team. Systematic review has confirmed that patients managed in closed units have lower mortality, fewer complications and shorter ICU and hospital stays.11 The vast majority of ICUs in the UK operate as closed units.
SURGERY 33:4
3
1:4e1:12
1:2
1:1
Table 2
181
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
National audit Input of data into national audit programmes is the cornerstone of any quality improvement programme.
unit’s mortality rate is vital if mortality is to be used as a valid outcome measure. Standardized mortality ratio (SMR) is the term applied to this ‘corrected’ measure of mortality:
Standardized Mortality Ratio ¼ Actual Mortality Rate=Expected Mortality Rate
Quality indicators e process
SMR allows comparisons of mortality figures between critical care units and within a single unit over a period of time. For example, if an SMR for a particular unit is more than 1, this would suggest that more people than expected were dying within that unit for its case mix and disease severity at presentation. This measure of worse outcomes for patients within a unit when compared with the expected mortality rate, potentially allows for the identification of areas of care in need of improvement. These could then be addressed and SMR re-assessed to confirm an actual clinical benefit to patients. Consequently, it becomes possible to monitor an individual department’s performance at any time and, in addition, to track its outcomes over time. SMR relies on an accurate prediction of mortality for a clinical presentation. For this we use critical care scoring systems.
Care bundles A care bundle is defined by the Institute for Healthcare Improvement as ‘a small straightforward set of evidencer-based practices, generally three to five, that when performed collectively and reliably, have been proven to improve patient outcomes’. Several care bundles are in common use including the ventilator care bundle, the central venous line bundle and the sepsis care bundle. The ventilator care bundle has four aspects: raising the head of the patients bed to 30e40 stress ulcer prophylaxis DVT prophylaxis spontaneous breathing trials. Use of this bundle has been shown to reduce the incidence of ventilator-associated pneumonia. The central venous line bundle is a set of five simple steps that have been proven to reduce the incidence of catheter-related bloodstream infection. The sepsis care bundle developed by the Surviving Sepsis Campaign is divided into early and late goals and has been associated with a reduction in mortality in a small cohort study.
Scoring systems Scoring systems are applied throughout clinical practice and critical care. They can be generic (applied to all disease presentations) or specific (applied to a single disease process). They can be anatomical (e.g. radiological grading of sub-arachnoid haemorrhage) or physiological (e.g. APACHE II). They can look at disease severity on admission to critical care or track the patient at regular intervals throughout the clinical course. An ideal critical care scoring system should be quick and easy to perform, involve minimal recording bias and have good discrimination and calibration. Discrimination is the ability of the scoring system to predict mortality in individuals without ambiguity. Currently all generic critical care scoring systems are poor discriminators and are unreliable when applied to individual patients when trying to predict an individual’s mortality. Calibration is the similarity between predicted outcome and actual outcome within a population. It is calibration that is important when using scoring systems to calculate SMRs and currently there are several critical care scoring systems that show an acceptable level of calibration and can be used to calculate SMR. The scoring systems used to calculate expected mortality rates and therefore SMR within a population are generic and are developed by prospectively collecting data sets from population groups where mortality is recorded. They apply a relative weighting to each patient variable, measure physiological derangement and take into account diagnosis and co-morbidities. Unfortunately, the validity of these scoring systems is degraded when they are developed on a population group other than the one to which it is being applied and over time as critical care practices and therapy change and develop. One of the most widely used scoring systems worldwide is the Acute Physiology and Chronic Health Evaluation score (APACHE II). This was developed by Knaus et al. and published in 1985. It used data from 5815 critical care admissions in the USA between 1979 and 1982. Despite widespread use its application to current
Early enteral nutrition The use of early enteral nutrition is associated with improved nutritional intake and trends towards reduced infectious complications and mortality. Several national practice guidelines recommend its appropriate use. End of life care pathways Both the Department of Health and the Robert Wood Foundation in the USA have defined quality markers for end of life care: patient- and family-centred decision-making communication continuity of care emotional and practical support symptom management and comfort care spiritual support emotional and organizational support for intensive care clinicians. In the UK the Liverpool Care Pathway is the most common tool incorporating these domains. Quality indicators e outcome Assessing and measuring outcome is a key aspect of evaluating the quality of critical care delivery. Mortality can be used as an easily definable measure of performance whereas morbidity measures can be more complex to define and interpret. Mortality Crude mortality measures are generally accepted as a poor measure of quality since they do not take into account variables such as case-mix, patient co-morbidities and disease severity. Correcting for these variables and therefore ‘standardizing’ a
SURGERY 33:4
182
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
receiving step down patients have minimal staff. Studies have shown they are associated with increased morbidity and mortality.
critical care patients in the UK could be questioned. Firstly, it was developed by analysing patient data from North America and secondly, these data are now over 30 years old. These criticisms, unfortunately, are limitations of many of the scoring systems within critical care.
Readmission rates Intensive care readmission rates range from 4 to 14%. Compared to patients with a similar severity of illness readmitted patients have twice the ICU length of stay and a two- to tenfold increased risk of death.
The Intensive Care National Audit and Research Centre (ICNARC) ICNARC was set up in 1994 alongside the ICS. Its goal was to ‘foster improvements in the organisation and practice of critical care within the UK.’ Currently more than 90% of adult general critical care units participate. The ICNARC model originally used data from over 200,000 admissions to the national case mix programme database and, with similarities to other scoring systems like APACHE, is now able to produce an expected mortality rate for a population. It combines a physiology score during the first 24 hours of critical care admission with patient demographics, reason for admission, co-morbid diagnoses, source of admission, surgical urgency and whether or not cardiopulmonary resuscitation (CPR) has taken place in the preceding 24 hours prior to admission. Its use for the UK population is valid since it has been generated using data from that population and it is calibrated regularly. The ICNARC case mix programme delivers each participant unit quarterly electronic data analyses reports. This report includes the SMR for the unit in addition to a myriad of morbidity data.
Healthcare-acquired infection Reducing the incidence of hospital-acquired infections has been a priority for the NHS as set out in the 2008/09 NHS Operating framework and the 2007 Public Service Agreement ‘Ensure better care for all.’ Mandatory surveillance has been in place since 2004, with a target reduction of 30% by 2010/11 compared with 2007/08 figures. Patients who have been in units for 48 hours or more and who screen positive are considered to have unitacquired infection. Long-term morbidity quality indicators In recent years there has been significant progress made in this previously under-explored area of long-term critical care morbidity outcomes. One-year survival following critical care in the UK stands at around 60%. It has been shown that amongst these survivors they continue to not only have a persistently raised mortality rate when compared with a matched population, but also suffer from increased rates of physical and psychological disease. The estimated prevalence of disease following critical care is: physical half of all critical care survivors have some degree of dependency in activities of daily living at 6 months 30% of survivors of working age will not return to work psychological anxiety 12e43% depression 10e30% symptoms of post-traumatic stress disorder 5e64%. However, quality trials in this area are still lacking and it remains unclear as to the exact prevalence of disease in this group of patients. Efforts are being made currently via the Intensive Care Outcome Network study (ICON) and the Intensive Care After Network (ICANUK) to recruit patients to try to answer this question in more detail.
Morbidity It is generally accepted that SMR is the key clinical outcome quality indicator for adult critical care. In recent years however, other outcome indicators have been explored as potential markers of quality. Short-term morbidity quality indicators Examples of these outcome measures would include: length of critical care admission ventilator days night-time discharges readmission rates non-clinical transfers healthcare-acquired infection rates incidence of unit-acquired meticillin-resistant Staphylococcus aureus (MRSA) bacteraemias incidence of all unit-acquired bacteraemias unit-acquired MRSA or Clostridium difficile incidence of ventilator-acquired pneumonia. Many of these markers are the source of public health, national or local targets (e.g. hospital-acquired infection rates); some are useful in order to collect data of incidence and prevalence of clinical conditions nationally (e.g. admission following CPR). Much of the data is collected and reported on behalf of UK critical care by ICNARC. However, data validating the use of many of these measures as a marker of quality are lacking. Currently, there is an assumption that these surrogate end-points potentially could be associated with long-term benefits.
Intensive care follow-up clinics Critical care follow-up clinics are becoming increasingly available in many hospitals. They provide opportunity for local audit of practice but also the opportunity of larger scale data collection that may facilitate future studies like ICON. Exactly what should be assessed and how is still a source of debate but standardization is beginning to occur. Longer-term outcome measures that may be assessed following critical care discharge include: physiological respiratory volumes (e.g. vital capacity) gas flow (e.g. FEV1) pulmonary diffusion capacity upper airway visualization (e.g. bronchoscopy)
Night time discharges Discharges from intensive care units during night shift hours are not planned, handover is frequently less than optimal and wards
SURGERY 33:4
183
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
physical functional status assessment of activities of daily living (e.g. Katz’s measurement) mental and neuropsychological functional status anxiety and depression (e.g. HADS survey) post-traumatic stress disorder (e.g. PCL-C questionnaire) health-related quality of life measures Short Form Health Survey (SF-36), Sickness Impact Profile scale (SIP) Perceived Quality of Life scale (PQOL) (38).
terms of length and quality of life. The cost of particular treatments can then be compared in quality adjusted life years (QALYs). A recent review of the cost-utility of intensive care treatment estimated it to be approximately £7000 per QALY which compares favourably with other acute care interventions.
Summary Crude and standardized mortality measures were the traditional markers of whether or not a critical care was achieving a satisfactory outcome for its patients. However, this has been increasingly shown not to be the whole story and markers of morbidity both in the short and long term as well as following up patients after critical care is increasingly being seen as a valid approach to the assessment of quality of care. If we are to improve outcomes for critical care patients we must improve quality in all three areas of structure, process and outcome using an organized Quality Improvement Programme. We must also be vigilant and adapt as markers of quality will change, as our speciality develops into the future. A
Satisfaction surveys Satisfaction surveys have long been seen as a potential measure of quality in healthcare services. Critical care provides additional challenges to this form of assessment since many patients will have altered consciousness and poor recollection of their critical care stay. In addition, a patient’s family often play a more prominent role in a critical care admission. Critical care physicians may frequently spend longer periods of time discussing clinical decisions and care plans with a patient’s relatives than with the patient themselves. Assessing a patient and their family’s satisfaction with the critical care service they received is felt to be of increasing importance when assessing overall quality of care. This satisfaction has been characterized into six general domains: assurance: the need to feel hope for a desired outcome information: the need for consistent, regular and realistic updates proximity: the need to be physically and emotionally near the patient support: the need for support services if required comfort: the need for comfortable surroundings and waiting areas benefit: the need to feel that the treatments are in the patient’s best interest. Many of these satisfaction domains have been validated when assessed via post-care questionnaire and are now a part of routine critical care follow-up.
REFERENCES 1 Department of Health. Quality critical care e beyond ‘comprehensive critical care’. London: Department of Health Publications, 2005. 2 Department of Health. High quality care for all: NHS next stage review final report. London: Department of Health Publications, 2008. 3 Department of Health. Equity and excellence: liberating the NHS. London: Department of Health Publications, 2010. 4 Committee on Quality of Health Care in America. Crossing the quality chasm: a new health system for the 21st Century. Washington, DC: National Academy Press, 2001. 5 Donabedian A. Evaluating the quality of medical care. Milbank Q 2005; 83: 691e729. 6 Intensive Care Society. ICS quality indicators 2010. Available at: http://www.ics.ac.uk/intensive_care.../standards.../ics_quality_ indicators. 7 The Joint Commission. National hospital quality measures e ICU. http://www.jointcommission.org/PerformanceMeasurement/ MeasureReserveLibrary/SpecþManualþ-þICU.htm. 8 Australian Council on Healthcare Standards. Australasian clinical indicator report 1998e2006. 2007. Ultimo, Australia: ACHC, http:// www.achs.org.au/pdf/Australasian_CIR_8th_Edition.pdf. 9 Kim MM, Barnato AE, Angus DC, Fleisher LF, Kahn JM. The effect of multidisciplinary care teams on intensive care unit mortality. Arch Intern Med 2010; 170: 369e76. 10 Kane SL, Weber RJ, Dasta JF. The impact of critical care pharmacists on enhancing patient outcomes. Intensive Care Med 2003; 29: 691e8. 11 Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. J Am Med Assoc 2002; 288: 2151e62. 12 BACCN, CC3N, RCN. Standards for nurse staffing in critical care. September 2009, http://www.baccn.org.uk/downloads/BACCN_ staffing_document.pdf. 13 Department of Health. Comprehensive critical care. London: Department of Health Publications, 2000. 14 Kane RL, Shamliyan T, Mueller C, Duval S, Wilt T. Nursing staffing and quality of patient care. Evidence Report/Technology assessment No.
Audit and quality improvement programmes In 2002 the National Institute for Clinical Improvement defined clinical audit as ‘a quality improvement process that seeks to improve patient care and outcomes through systematic review of care against explicit criteria and the implementation of change’. The audit cycle involves identification of the problem, defining a standard, data collection, comparing performance against that standard and implementing change. The cycle is then repeated to assess the response to change. Application of the audit cycle to the aforementioned quality indicators can be used to improve patient outcome and is known as a Quality Improvement Programme.
Economic performance Measuring clinical performance cannot be completely divorced from measuring economic performance, which is achieved by the use of cost-utility analysis. This specific form of cost-effectiveness analysis measures cost in the form of monetary units but benefit in
SURGERY 33:4
184
Ó 2015 Elsevier Ltd. All rights reserved.
CRITICAL ILLNESS AND INTENSIVE CARE: I
15 Rhodes A, Moreno RP, Azoulay E, et al. Prospectively defined indicators to improve the safety and quality of care for critically ill patients: a report from the Task Force on Safety and Quality of the European Society of Intensive Care Medicine (ESICM). Intensive Care Med 2012; 38: 598e605.
151 (Prepared by the Minnesota evidence based practice Center under Contract No. 290-02-0009.). March 2007. Rockville, MD: Agency for Healthcare Research and Quality. AHRQ Publication No. 07-E005, http://www.ahrq.gov/downloads/pub/evidence/pdf/ nursestaff/nursestaff.pdf.
SURGERY 33:4
185
Ó 2015 Elsevier Ltd. All rights reserved.