Portal de Periódicos da CAPES

Health services research and anaesthesia

2012; Wiley; Volume: 68; Issue: 2 Linguagem: Inglês

10.1111/anae.12063

1365-2044

Michael P. W. Grocott, Michael J. Galsworthy, Ramani Moonesinghe,

Hip and Femur Fractures

Health Services Research (HSR) is critically important to the development of anaesthesia as a specialty for three reasons: first, there is a paucity of high-quality clinical data against which to benchmark clinical practice; second, there are the limitations to the clinical measures available to us; and third, we must consider the often complex and confounded nature of the interventions that we seek to evaluate. This notion is not novel: a similar conclusion was recently reached by the UK Intensive Care Foundation, which identified HSR as one of three areas of strategic research focus 1. In fact, the UK's Intensive Care National Audit & Research Centre (ICNARC: see https://www.icnarc.org/) is already a global leader in applying HSR to critical care. Anaesthesia is the largest hospital speciality within the National Health Service, and yet, arguably, our profession is served by the poorest quality data of any discipline. This is a remarkable paradox, given the time and effort invested by individual anaesthetists intensively recording individual patients' observations during anaesthesia. Sadly, in most cases these data immediately become practically unavailable for systematic evaluation. At the same time, the epidemiological data we need at a specialty level is of limited breadth and depth, and difficult to access. By contrast with our surgical peers, who can call on data from a number of high-quality, condition-specific, national clinical audits such as the National Bowel Cancer Audit (see www.ic.nhs.uk/bowel) and National Hip Fracture Database (see http://www.nhfd.co.uk/), we have very little with which to benchmark care and outcome. Bluntly, we are more than 25 years behind the cardiothoracic surgeons, and we need to catch up fast. It is ironic that our specialty counts John Snow, the founding father of clinical epidemiology, amongst its alumni. Health Services Research is the study of the provision of healthcare to the population. Unlike public health research or health economics research, HSR is focused on the institutions that provide healthcare and their services. It examines their processes and outcomes, harnessing economic, organisational, statistical and medical knowledge in order to evaluate scientifically the quality and efficiency of the healthcare service being provided. Health Services Research is the glue that should bind clinical practice to developments in basic medical science. As well as assessing the treatments that have made it from the bench to the bedside, it should examine what is missing at the bedside in order to inform the bench. Ultimately, HSR is simply defined as research with the aim of improving patient outcomes at better cost. To clarify the relationship of HSR with other areas of research, we note that it conceptually encompasses ‘classical’ descriptive clinical epidemiology as well as research with a focus on the effectiveness of healthcare interventions, including late-stage clinical trials and health technology assessments. The boundary between HSR and what is now referred to as ‘experimental medicine’ (the patient end of the bench-to-bedside innovation pathway) is blurred (Fig. 1). However, whereas studies such as clinical trials may evaluate efficacy under ideal conditions, HSR more usually focuses on the effectiveness of the treatment within the complex real-world setting. This often means considering the impact of health systems factors beyond simple patient-intervention interactions. Characteristically, HSR evaluations are of complex innovations such as care pathways and organisational changes. This also means that the ‘real’ data are often retrospective and messy, in that the patient allocations are confounded – rather than the randomised and blinded studies that one would preferably design. It is a characteristic of HSR that studies may involve both mixed (quantitative and qualitative) and multiple methods. For example, a clinical trial of the introduction of a new care package in emergency surgery might involve semi-structured interviews with practitioners, consensus development through a Delphi process, implementation science, health economics and physiological measurement, as well as expertise in trial design and conduct from a clinical trials unit. The important point is that all evaluations of treatments and processes are viewed within the context of the healthcare system. Although the HSR label may be relatively new within the speciality of anaesthesia, the academic discipline of HSR is well established within our major universities and national funding streams, and HSR researchers have a significant impact on national healthcare policy. Recent landmark HSR publications relevant to anaesthetic practice include the evaluation of the WHO Surgical Safety Checklist by Atul Gawande's group 2, Shukri Khuri's report of the relationship between short-term postoperative complications and long-term mortality 3 and Amir Ghaferi's ‘failure to rescue’ papers examining variation in hospital mortality following major surgery 4, 5. Notable contributions in the UK context include Rupert Pearse's description of the high-risk surgical population in the UK 6, the risk-adjusted comparison of mortality rates after surgery between patients in the USA and the UK by Monty Mythen's group 7, and the recent publications of audit data from the Hip Fracture Perioperative Network 8 and the Emergency Laparotomy Network 9. Studies evaluating the cost and ‘profitability’ of peri-operative care are of increasing interest in an environment where the NHS is under pressure to improve efficiency while maintaining safety and effectiveness 10, 11. Large, high-quality datasets are the foundations on which much of HSR is built. Strictly, the systematic collection of data about healthcare within databases, registries and audits is not a research activity: there is often no specific research question being addressed. However, the data obtained permit novel observations of clinical phenomena (e.g. the relationship between short-term and long-term outcomes) as well as providing important information on feasibility and design, e.g. for sample size calculations for interventional studies. As a consequence, HSR researchers are often central to the development and management of clinical audits. Some clinical audits have developed out of small observational studies; for example the ICNARC Case Mix Programme developed from the initial UK validation of a US-developed critical care scoring system 12, 13. High-quality clinical audits incorporate descriptions of structure, process and outcome to provide a comprehensive picture of the impact of healthcare delivered to the patients. Within the UK, outcomes are currently classified according to the ‘Darzi’ domains of safety, effectiveness and experience 14. Casemix adjustment is a vital element of any analysis in order to account for the effect of variation in patient-specific input variables (e.g. age, comorbidities, socio-economic status) on output variables (clinical outcomes), and thereby separates out the effect of variation in delivered care. However, it is important to be aware that casemix adjustment may be imperfect due to the effect of residual confounding from unmeasured and unanticipated variables: a case of the ‘known and unknown unknowns’. The data within such audits are critically dependent on the quality of the clinical measures used. The types of measures available and the methods underlying the development and validation of such measures are beyond the scope of this article, but are covered in depth elsewhere 15. Important characteristics to consider are reliability, validity and the burden imposed on data collectors. Some of these measures may also have utility in clinical practice. For example, Clinical Risk Prediction Rules (e.g. those including ASA physical status 16 or the Surgical Risk Scale 17) may have dual utility in that they not only retrospectively provide casemix adjustment within a clinical audit, but also may prospectively characterise risk before surgery and thereby guide subsequent management and communication with patients. However, this dual role can introduce a paradox whereby the evaluation of a casemix adjustor may be confounded by the changes in practice induced by clinicians' knowledge of the measures: so-called ‘confounding by indication’ 18. In anaesthesia, these clinical measures are in their infancy: there are limitations in both the breadth and depth of coverage. We have major deficiencies in casemix/risk and outcomes measures in several areas of anaesthetic clinical practice and the development and validation of the measures to fill this unmet need are important research priorities within our profession. So what will the collection and analysis of such data achieve? First, it will allow rational management of resource-intensive structures and processes that currently function in a relative information vacuum. Second, there is a substantial evidence base supporting the notion that, in and of itself, collecting data seems to improve outcome 19. Third, such data provide a vital benchmarking resource to facilitate the identification of best processes and outcomes, thereby driving convergence towards best practice by all providers. Fourth, they provide the infrastructure to support both quality improvement initiatives and primary research, thereby driving progress in the development of new and better standards of care. Finally, and importantly, they provide high-quality information to patients and doctors, enabling shared decision making of the highest quality 20. Such data allow the weighing of competing harms, for example the likelihood of short-term postoperative harm following elective aortic aneurysm repair weighed against the long-term likelihood of aneurysm rupture. Whilst such decisions are often framed as binary (death from surgery vs cancer survival), improved epidemiological data allow better communication of the range of possible outcomes, including prolonged hospital stay and ongoing physical and mental impairment. Limitations in the epidemiological data describing anaesthetic practice are also linked to scepticism about the magnitude of harm following surgery. This scepticism is of concern because it is a commonly voiced reason for not innovating or adopting evidence-based interventions, e.g.: “I do not see that level of morbidity/mortality in our practice”. The landmark publication of the European Surgical Outcomes Study (EuSOS) provides valuable data in this respect 21. The UK was the largest contributor to EuSOS and enrolled around 50% of eligible patients during the recruitment week in April 2011. However, whilst these data are an invaluable resource, there is little substitute for ongoing systematic data collection within a national registry/audit. The Royal College of Anaesthetists' (RCoA's) A National Strategy for Academic Anaesthesia (‘Pandit Report’) of 2005 led directly to the development of the National Institute of Academic Anesthesia (NIAA) 22. The NIAA was constituted in order to develop the academic profile of anaesthesia nationally, to facilitate high-profile influential research and to support training and continuing professional education in academic anaesthesia (see http://www.niaa.org.uk/). The funding partners of the NIAA are the Association of Anaesthetists of Great Britain and Ireland, the RCoA, the journals Anaesthesia and the British Journal of Anaesthesia, and the anaesthetic specialist societies (e.g. Difficult Airway Society). Upon this foundation, the NIAA Health Services Research Centre (HSRC) was established to address the unmet need for coordinated HSR within anaesthesia (see http://www.niaa-hsrc.org.uk/). The HSRC remit encompasses anaesthesia, peri-operative medicine, pain and the anaesthetic subspecialties, but not critical care as ICNARC meets that need within the UK. The aims of the HSRC are to initiate and coordinate large clinical audits relevant to the practice of anaesthesia, to lead innovation in clinical measurement methods, and to facilitate and conduct research for patient benefit. Subsidiary aims include contributing to the training of HSR researchers within anaesthesia and helping to develop the portfolio of major UK clinical anaesthetic studies. We seek to achieve these goals through an open and collaborative model of working and welcome approaches from anyone with an HSR challenge or with an interest in becoming involved. Current HSRC projects include a programme of systematic reviews of risk and outcome measures, a national survey recording the use of such measures, and a working group to synthesise these inputs into recommendations for clinical departments. An early success has been the commissioning of the National Emergency Laparotomy Audit by the Health Quality Improvement Partnership (HQIP) (see http://www.hqip.org.uk/), the result of a productive collaboration between the HSRC and the Emergency Laparotomy Network (see http://www.networks.nhs.uk/laparotomy). We are also collaborating with several groups on developing clinical trials. Finally, the HSRC has also taken on the oversight of the National Audit Projects (NAPs; see http://www.nationalauditprojects.org.uk/). The NAPs are distinct from the clinical audits in that they are discrete fixed-duration projects (rather than ongoing audits) focusing in general on the incidence and consequences of rare phenomena (e.g. accidental awareness under anaesthesia). The aim of the NAPs is to identify factors that can be used to improve patient care and they are usually conducted in partnership with the relevant specialist society. Future HSRC work may focus on evaluating the impact of NAP reporting in routine clinical practice, and developing strategies to improve and evaluate the implementation of NAP recommendations. Perhaps the most renowned public health physician of the modern era is Avedis Donabedian, who first described the ‘structure-process-outcome’ model that is used to describe and evaluate the delivery of healthcare 23. Despite his pivotal role in developing health services research, he also recognised the importance of individual clinician engagement in providing quality healthcare: “Systems awareness and systems design are important for health professionals, but are not enough. They are enabling mechanisms only. It is the ethical dimension of individuals that is essential to a system's success. Ultimately, the secret of quality is love.” It could be argued that anaesthesia as a profession has no shortage of ethics or love, but is limited by lack of systems development and evaluation. Looking forward, we should aspire to see comprehensive audit of peri-operative care, using reliable, validated and precise measures 24. We will achieve this through a combination of anaesthetic-led audits focusing on peri-operative care (such as the National Emergency Laparotomy Audit) and the collation of relevant data from established condition- and procedure-specific audits (e.g. National Bowel Cancer Audit, National Adult Cardiac Surgery Audit (see http://www.ucl.ac.uk/nicor/audits/Adultcardiacsurgery)). We should use such data to benchmark all aspects of our practice against the very best achievable, nationally and internationally, and to drive quality improvement within our specialty. Finally, we need to use the tools of HSR to explore important research themes in anaesthesia and peri-operative medicine: defining more precisely the relationship between short- and long-term outcomes 3; exploring the basis of ‘failure to rescue’ in the peri-operative context 4; and understanding the pathophysiological basis of postoperative harm through better understanding the contributing risk factors and processes 25. In this way, HSR can be a fundamental tool in our common aim: to improve the safety and effectiveness of care, and the quality of experience we deliver to our patients. MPWG holds the British Oxygen Company Chair of Anaesthesia at the RCoA. MPWG is funded in part by the University Hospitals Southampton NHS Foundation Trust – University of Southampton Respiratory Biomedical Research Unit which received a portion of its funding from the UK Department of Health's National Institute of Health Research (NIHR) Biomedical Research Unit funding scheme. SRM is funded in part by the University College London Hospital–University College London Biomedical Research Centre which received a portion of its funding from the UK Department of Health's NIHR Biomedical Research Centre funding scheme. MPWG is Director and MJG Health Services Researcher, NIAA HSRC. MPWG, MJG and SRM serve on the Executive Board of the HSRC. MPWG and SRM serve on the Board and MPWG serves on the Research Council of the NIAA. MPWG and SRM have received funding from the NIHR, the NIAA and the Frances and Augustus Newman Foundation to conduct HSR. The views expressed are those of the authors and not necessarily those of the bodies named.