|
| ||
|
| | Annex A > Chapter 18 - Medical and Clinical Audit > Audit: the national perspective > The setting of standards << previous | next >> The setting of standards74 Central to the concept of audit, as defined at para 5 above, was the idea that standards of clinical care should first be set; then performance assessed, and possible improvements in practice identified and implemented. `In the middle and late 80s there was developing interest in more direct and locally based medical audit in which individuals and groups of clinicians would define the standards that they wish to achieve, compare their actual practice with those standards, and institute remedial action where the standards were not being achieved, re-auditing performance subsequently to ensure that the remedial actions had been successful. This process became known as the Audit Cycle and forms the basis of all subsequent medical and clinical audit. In the year before the publication of "Working for Patients", the Department funded a number of the Medical Royal Colleges to develop medical audit projects on this basis. The then Chief Medical Officer also secured the endorsement of all College Presidents that such activity should be an integral part of routine clinical practice. However, at that stage medical audit was very much a minority activity pursued by enthusiasts.' [86] Increasing pressure developed for doctors' clinical activity to be included in NHS initiatives concerning quality. For example, evidence emerged about unexplained variations in practice related to length of stay, hospital admission rates and variations in outcome. A number of arguments about hospital clinical competence were well publicised. There was an increase in the willingness of pressure groups to publicise information about substandard services. [87] 76 Professor Sir George Alberti, President Royal College of Physicians (RCP), pointed to the difficulty of measuring quality of care and outcome of care and said that comparative information evidencing national standards did not start to emerge until after 1990. [88] 77 Counsel to the Inquiry referred Sir Graham Hart, NHS Management Board Director of Operations from 1985 to 1989 and from March 1992 to 1997 Permanent Secretary at the DoH, [89] to the 1983 `NHS Management Inquiry Report' in the following exchange: `Q. In the Griffiths report - we will just have a look at some of the general comments which he made ... This comes from Griffiths, it is page 10 of what is acknowledged to be a short but effective report. In paragraph 2, under his general observations, he describes the NHS not having a profit motive but being enormously concerned with the control of expenditure: "Surprisingly, however, it still lacks a real continuous evaluation of its performance against criteria such as those set out above ... Rarely are precise management objectives set. There is little measurement of health output. Clinical evaluation of particular practices is by no means common and economic evaluation of those practices extremely rare." `Leaving aside the economic and leaving aside the question of output, the number of operations done, clinical evaluation of particular practices is by no means common. `In this paragraph as a whole, what Griffiths appears to be observing, and the implication is, complaining about, is that the NHS had no proper measurement of the quality of the care it was providing in general terms. `First of all, from your own perspective, was he probably right about that, at the time?' `A. Yes. I mean, I would say, I think, what he was saying was that there was no system, if you like. Some of these things happened, but they did not happen in an organised and systematic way. I think that is true. He was spot-on, there.' [90] 78 The opinion of Professor Sir Kenneth Calman, Chief Medical Officer (CMO) for England 1991-1998 [91] on the issue was explored in the following exchange: `Q. ... Sir Graham Hart ... has told us that, throughout the period of particular concern to this Inquiry, there was no proper measurement of the quality of care which was available within the NHS, looking at the question of the delivery of care by hospitals. `A. No, I do not think that would be my view, because for really a very long time, the outcome of the health care has been part of the responsibilities of individual doctors and indeed trusts and before that, hospital boards. It would be impossible to manage a system without knowing what the outcome was. That was done in a variety of different ways over the years, but I think in terms of the outcomes of healthcare, there are difficulties in measuring sometimes the outcome of health care. Mortality is a very relevant way to measure, but once you move into other areas like quality of life, for example, it becomes more difficult to measure, but in terms of the outcome of healthcare, 30-day mortality, wound infection rates have been recorded and reported for a very long time.' [92] 79 Sir Barry Jackson said that the: `... setting of the standards have created considerable problems in many areas, and in 1989/1991 those standards in most instances were not recognised; therefore to all intents and purposes, they did not exist.' [93] 80 Dr John Roylance, Chief Executive of UBHT 1991-1995, was asked about a paediatric cardiology report of March 1992, set out upon the MAC standard form. The `audit' topic reviewed was paediatric cardiac surgical mortality for 1991. The document contained comparisons with previous years. [94] Dr Roylance gave evidence that he regarded this kind of exercise not as audit but as a review of recent outcomes. [95] 81 Looking at the same document, Sir Barry Jackson agreed with this emphasis upon the centrality of standard-setting: `In the strict meaning of the term, I would agree with Dr Roylance, as I said earlier, because there is no standard set there with which to compare the mortality other than previous years, but there is no acceptance written there that the previous years' figures are the standard to which they were judging the current years' standard.' [96] 82 Sir Barry Jackson further gave evidence that if the standard used was in the form of national indicators of outcomes, drawn, for example, from the cardiothoracic surgeons' register, this was not audit in the formal sense, unless: `... it has been defined and agreed initially that that is the standard to which one is aspiring ...' [97] `I believe Mr Jackson is correct when he has asserted that clinical audit is strictly concerned with setting standards and then auditing activity against those standards. However, there are many preliminaries to a clinical team reaching the stage where they can set standards in an authoritative way and then measure their activity correctly. It is part of the quality assurance concept for providers and commissioners of services to observe that this process is taking place.' [98] 84 Dr Morgan further commented on Dr Roylance's evidence: [99] `Dr Roylance's view in his statement is that true clinical audit was happening rather infrequently and, of course, this depends on the definition of audit. The early definition spoke of systematic, critical appraisal of clinical activity and includes case note review by peers, etc. This kind of activity was not uncommon throughout the Trusts, including the UBHT. However, if one uses a more modern definition of clinical audit - the explicit setting of standards and vigorous measurement of activity to assess the extent to which its standards have been met - then Dr Roylance is right.' [100] The nature of audit85 In 1990 the Standing Medical Advisory Committee (SMAC) wrote: [101] `The essential nature of medical audit is a frank discussion between doctors, on a regular basis and without fear of criticism, of the quality of care provided as judged against agreed standards ... It should lead to action where practice has not matched the agreed standards so that the quality of medical care is improved. The principles of medical audit can be compared with those of feed back loop control in which the expected standard of care is defined in whatever terms are agreed to be appropriate, reality is compared with the defined standard and practice is changed in the light of this comparison. This is referred to as the "audit cycle". `Although sharing similar objectives with medical audit in respect of medical education and training, the "grand round" or "interesting case" type of clinical meeting does not meet the requirements of medical audit. Medical audit is a systematic structured procedure with the express purpose of improving the quality of medical care. Wherever possible it should be quantified. `Medical care can be considered in terms of structure, process or outcome. Structure is concerned with the amount and type of resources available, for example the condition of buildings, the number of beds available and staffing levels. These are easy to measure but are not necessarily good indicators of the quality of care provided. Process relates to the amount and type of activity expended in the care of a patient. Unless resources are severely limited process has more significance than structure and in many circumstances it is the only measure available. The most relevant indicator of quality of care is outcome. ... Examples of outcome measures include mortality, such as perinatal mortality and perioperative deaths, residual disability, relief of symptoms and patient satisfaction ...' Types of audit`There are two main approaches to the practice of medical audit. They are (i) retrospective internal audit within a specialty, hospital, general practice or district community in which records are used to review past events, and (ii) concurrent audit which is a continuous assessment of patient management. In both types of audit results are compared with agreed standards, which may be implicit or explicit, protocols or criteria. We feel that retrospective internal audit is likely to be the most appropriate approach for the introduction of medical audit but these approaches to audit are not mutually exclusive. `Ideally the basis of audit should be outcome but in practice it is often not. Usually audit of "process" is carried out on the assumption that good process gives rise to good outcome. The subject of audit may include administrative processes (such as medical records, referral and discharge letters), clinical processes (use of drugs, investigations and procedures), clinical condition (classified by diagnostic category) or outcome (return to work, ambulation or unexpected death). `Medical audit is now increasingly recognised as a component of medical practice and therefore all doctors should be expected to take part. The main components in the process are:
`Follow-up action is an absolutely essential feature of medical audit without which the justification for medical audit is lost. Medical audit should lead to changes in the organisation and availability of services, clinical policy and clinical practice with consequent improvement in the quality of medical care as measured by appropriate indicators.' [102] 87 Further, as a national policy upon the introduction of audit developed, guides to the process of audit began to be published. [103] 88 The Quality of Practice Committee of the RCA noted: [104] `Almost any medical activity may be usefully subjected to audit. Included under this heading are: `Provision and use of specific services (e.g. operating theatre time, ITU, pain services, etc.) 89 Dr Baker, in his evidence, noted the debate between the UBHT and the District on the role of measures of process and of outcome. The UBHT wished to concentrate on measures of outcome. [105] He observed: `By the end of 1995 there was recognition within the medical literature and amongst professional and health service organisations that audit carried out productively to benefit patients was an exacting task, no less so than any other quantified approach to the measurement of quality or resolution of uncertainties. The National Centre for Clinical Audit published "Good Practice in Clinical Audit" in 1996 which summarised in particular the difficulties of audit of clinical outcome and encouraged audit of processes of care in relation to explicit criteria. Auditing clinical outcomes requires essentially that cause and effect are well understood in relation to the contributory components of healthcare and their actual relationship to variations in outcomes. In most instances of healthcare this relationship is not well understood.' [106] The effectiveness of the national audit programme90 The Inquiry received evidence upon the successes and failures of the Government's introduction of medical and, subsequently, clinical audit. 91 The Inquiry was also referred [107] to a number of research studies that had been undertaken to assess the impact and effectiveness of the national audit programme. [108] 92 The case study of the implementation of audit in general medicine in four hospitals undertaken in 1991/92 on behalf of the King's Fund [109] found that:
93 The Clinical Accountability Service Planning and Evaluation (CASPE) [110] study of the impact of the medical audit programme [111] surveyed provider units towards the end of 1993. It found:
The primary barriers to the development of audit, according to the study, may be summarised as follows: [113]
94 The CASPE study also found that by the end of 1993, clinical, as opposed to medical, audit was not well established: `It seems that medical audit has become a part of the fabric of practice for almost all medical staff. It would be difficult to find many doctors in the HCHS [Hospital and Community Health Services] whose working life has not been touched in some way by audit over the last four years. While this does not mean that medical staff are all committed to audit, or involved in assessing the quality of their own practice, it is a considerable achievement. Among other clinical professions - such as nurses, therapists, pharmacists, scientists and others - participation is probably much lower. This is not necessarily because members of those professions have not wanted to be involved - they may well exhibit the same spectrum of opinion as medical staff; from enthusiasm to disinterest in audit. Rather, it is because the medical audit programme was led by doctors and was focused on securing the involvement of medical staff - the involvement of other professional groups has often not been welcomed or encouraged. Indeed, enabling these much larger and more numerous professional groups to take part in clinical audit in the future presents some real challenges.' [114] 95 The study noted that: `... for the many clinicians who were participating in audit for the first time, the prospect of sharing potentially sensitive information with colleagues from other disciplines was not appealing, especially at first.' 96 `Evaluating Audit: Provider Audit in England: A review of twenty-nine programmes' [115] set out seven `critical success factors' for clinical audit programmes. These were: `Clinical Leadership This seemed to be the most important single determinant of an audit programme's success. `Vision, strategy, objectives and planning Providers with successful audit programmes had an explicit vision of what the audit programme was there to do, which had been communicated to everyone and was kept to consistently. `Audit staff and support Successful audit programmes had good audit staff who were recognised as an expert resources for advice and support and valued as important members of the team. `Structures and systems Many audit programmes faltered because they lacked basic structures and systems, e.g. for managing the workload, prioritising, timetabling, monitoring and reporting. `Training and education Few providers had recognised the need for training in audit skills which, despite their professional background, many clinicians did not already possess. `Understanding and involvement As well as good communication, training and leadership, successful participation in audit programmes also depended on resources, time and appropriate incentives and sanctions. `Organisational environment Well-managed providers with good personal and professional relationships among staff and with purchasers were able to establish better audit programmes. Dysfunctional organisations with a history of internal and external conflict and dissent found establishing audit more difficult. Thus the organisations likely to be most in need of audit and quality improvement were probably the least able to make it happen.' 97 In oral evidence to the Inquiry, Dr Walshe criticised the professional guidance from the DoH on the earlier approach of medical audit: `I think it would be true to say that the Department's proposals for medical audit in the NHS at that point in time could be criticised with hindsight as not being particularly directive, not if you like mandating a particular process, not requiring the organisations to undertake audit in a particular way and for also perhaps not putting in place particularly strong incentives or requirements for people to engage in this process.' [116] 98 Professor Sir George Alberti told the Inquiry [117] that it appeared that the DoH's focus was more on throughput and waiting lists than on outcome or quality of care and that the lack of guidance given in the area of audit was a reflection of this: `They were not interested in results; they were interested in as many people passing through the system as possible for as low a cost as possible ... commercial considerations did seem to enter into it rather strongly.' [118] 99 Dr Ernest Armstrong, the Secretary of the British Medical Association (BMA) from 1993 to date, took the view that audit, be it medical or clinical, had not been successful thus far. He said that evidence showed: `... clinical audit has not actually delivered the results that early enthusiasts, and I include myself amongst those, might have expected. We still have a long way to go to change the culture to allow doctors to take part in an open and responsive way in a supportive managerial structure that will ensure that we are not in a punishment mode; that when we find things not as they should be, we do not punish them [sic]; we have to put them right.' [119] 100 He said that the evidence also showed that the BMA had encouraged doctors: `... to take part in medical audit, in clinical audit, to discuss with peers, not only with medical peers but actually recognising that this involves discussing with peers in the wider health care team, the outcomes of their work, in a situation where, of course, as you would expect, people who do not have a problem turn up and people who do have a problem do not, and do not take part. `The question is, how does one encourage people to learn that by taking part they can only benefit, that this is not a threatening or censorious procedure, it is a learning exercise for everyone: one in which the aim is to generate support for something which is not as good or not at the standard that it was supposed to be and to generate a method of putting it right so that the next time you audit it, it is where it is supposed to be? `That is very difficult and it is particularly difficult if doctors think that by talking frankly and fully and openly with their colleagues about just why they are not at the standard, the outside standard, however it is measured, that they ought to be, the result is going to be some kind of disciplinary action ... and then one, I think, should be able to understand the reluctance of doctors to take part and the need for the BMA in doing as I said earlier this morning, its role of describing what leading edge looks like and where people ought to be in advocating doctors to move closer to the leading edge.' [120] The constraints (if any) placed on confidentiality and/or the assurance of anonymity [121]101 The implementation of audit in the late 1980s and early 1990s and, in particular, the development of information technology systems to support it created an accessible collection of data relating to the performance of individual clinicians. The perception was that this information was capable of misinterpretation and was potentially damaging both to individual clinicians and to public confidence in the healthcare system. [122] 102 Clinicians were concerned that data collected could be disclosed to patients or to patients' representatives in court actions for clinical negligence. They were also concerned about disclosure to non-professionals or managers, who might misuse it for `whatever purpose'. [123] 103 In 1990, SMAC wrote: [124] `Concern has been expressed that any record of the discussions of a medical audit meeting could be subject to legal subpoena. It is important that doctors should not feel that they are under a greater threat of litigation because of their involvement in medical audit. Confidentiality is essential. We recommend therefore that the documentation of audit meetings are [sic] provided in an appropriately anonymised form so that the general conclusions of the meeting and recommended action are recorded while the cases used in the discussion are not in any way identifiable.' 104 In May 1991, the Quality of Practice Committee of the Royal College of Anaesthetists advised: `In common with other Colleges and Faculties, the College of Anaesthetists has considered the medico-legal consequences of audit. When data are collected centrally every attempt is made to render its [sic] origin unidentifiable and to destroy secondary records as soon as possible. `Nevertheless, it is the responsibility of all clinicians to keep records of events which involve patients, and it is an offence to alter or destroy such records. The College has been led to understand that all primary records (case notes, anaesthetic records, etc.) are ultimately accessible to patients' relatives and their legal representatives. Secondary data extracted from such records can be rendered anonymous and destroyed. `This should not deter clinicians from their responsibilities for performing audit, although it should cause great care to be taken when an opinion is given and recorded as to the cause of any untoward event which may be discussed under the heading of morbidity and mortality.' [125] 105 Although in Dr Walshe's and Sir Barry Jackson's opinion concerns about confidentiality appeared to have waned over time, [126] these concerns were prevalent at the time of the formal introduction of audit in 1990 and, in the opinion of Mr Wisheart, until around 1995. [127] 106 Dr Thomas told the Inquiry that in response to these initial concerns, guidelines and protocols on confidentiality were contained in DoH Working Paper No 6 [128] and the 1991 recommendations were contained in the RHA's protocol on confidentiality. [129] 107 Dr Roylance believed that this document, `Confidentiality of Clinical Audit Information', was in response `... to the concerns of many doctors about the potential access to audit information by managers and it was agreed in Bristol that any requests for audit information, whether from managers or from purchasers, should be channelled through the appropriate Clinical Director.' [130] 108 Dr Walshe told the Inquiry: `... Data about individual clinicians would stay within the team and the Clinical Director, so that the Clinical Director had a key role there. That did not mean that if that Clinical Director had concerns about a particular individual, they would not then be able to raise those concerns, and indeed, they would have a duty to raise those concerns with those higher up in the organisation. But it was our kind of pragmatic response to try and find a middle way between the concerns of clinicians and the effectiveness of having an effective audit process.' [131]
Footnotes [86] WIT 0331 0003 Dr Winyard, Medical Director of the NHS Executive and Deputy Chief Medical Officer from 1993 to 1998 [87] INQ 0011 0008; `Medical and Clinical Audit in the NHS' [88] T9 p.43 Professor Sir George Alberti [89] WIT 0040 0001 Sir Graham Hart [91] WIT 0336 0001 Professor Sir Kenneth Calman [92] T66 p.5 Professor Sir Kenneth Calman [93] T28 p.92 Sir Barry Jackson [94] UBHT 0061 0161; paediatric cardiology report 1992 [95] T25 p.42 Dr Roylance. He also made the point that audit as he understood it was not only about measuring morbidity and mortality rates for surgical procedures and the like, but also included, for example, the monitoring of the effectiveness of equipment [96] T28 p.91 Sir Barry Jackson [97] T28 p.91 Sir Barry Jackson [98] WIT 0307 0018 Dr Morgan [100] WIT 0307 0014 Dr Morgan [101] `The Quality of Care' , report of the Standing Medical Advisory Committee, DoH, 1990, paras 4.1-4.3 [102] `The Quality of Care', report of the Standing Medical Advisory Committee, DoH, 1990, paras 7.1-7.2.1 [103] See, e.g., Shaw C. `Medical Audit - a Handbook', London: King's Fund, 1989 [104] WIT 0065 0596 Professor Strunin, May 1991 [106] WIT 0074 0040 Dr Baker [107] Walshe K and Ham C. `Acting on the evidence: progress in the NHS', NHS Confederation, 1997 [108] These included: Buttery, Walshe, Coles, Bennett. `Evaluating Medical Audit: The development of audit - Findings of a national survey of healthcare provider units in England', CASPE Research, 1994; Morrell C, Harvey G, Kitson A. `The Reality of Practitioner-Based Quality Improvement: A Review of the Use of the Dynamic Standard Setting System in the NHS of the 1990s', National Institute for Nursing, 1995; Willmot, Foster, Walshe, Coles. `Evaluating Audit: A review of audit activity in the nursing and therapy professions - findings of a national survey', CASPE Research, 1995; Buttery, Walshe, et al. `Evaluating Audit: Provider Audit in England: A review of twenty-nine programmes, CASPE Research, 1995; National Audit Office. `Clinical Audit in England', 7.12.95. Further evaluative studies were assessed in the Inquiry paper on `Medical and Clinical Audit in the NHS', INQ 0011 0016 [109] Kerrison S, Packwood T, Buxton M. `Medical Audit: Taking Stock. London: King's Fund, 1993; T62 p.3 Dr Walshe [110] CASPE is an organisation that undertakes research into audit mechanisms for a variety of organisations, including the DoH; T62 p.4 Dr Walshe [111] Buttery, Walshe, Coles, Bennett. `Evaluating Medical Audit: The development of audit - Findings of a national survey of healthcare provider units in England', CASPE Research, 1994 [112] Buttery, Walshe, Coles, Bennett. `Evaluating Medical Audit: The development of audit - Findings of a national survey of healthcare provider units in England', CASPE Research, 1994, p. 1-2 [113] Buttery, Walshe, Coles, Bennett. `Evaluating Medical Audit: The development of audit - Findings of a national survey of healthcare provider units in England', CASPE Research, 1994, p. 101-7 [114] Buttery, Walshe, Coles, Bennett. `Evaluating Medical Audit: The development of audit - Findings of a national survey of healthcare provider units in England', CASPE Research, 1994, p. 105 [115] Buttery, Walshe, Coles, Bennett. `Evaluating Medical Audit: The development of audit - Findings of a national survey of healthcare provider units in England', CASPE Research, 1994 [117] T9 p.42 Professor Sir George Alberti [118] T9 p.42 Professor Sir George Alberti [121] In this section the term `confidentiality' refers to the basis on which information may be made available which identifies individuals caring for a patient; that is, confidentiality in the context of data that refers to individual clinicians and clinical teams. Confidentiality, in the sense of protecting patients from being identified, was not a contentious issue in the Inquiry although it was an exercise that needed to be carried out to enable the use of data for audit. Dr Walshe confirmed that anonymising patient details was not a bar `to producing effective medical or clinical audit' [122] WIT 0323 0031 Dr Thomas [123] T62 p.19 Dr Kieran Walshe; T14 p.104 Professor Strunin [124] `The Quality of Care' , report of the Standing Medical Advisory Committee, DoH, 1990, para 8.5.2, p. 20 [125] WIT 0065 0599 Professor Strunin [126] T62 p.20 Dr Walshe; T28 p.96 Sir Barry Jackson [128] UBHT 0052 0306; DoH Working Paper No 6 [129] WIT 0323 0027 Dr Thomas; T62 p.121 Dr Thomas [130] WIT 0108 0046 Dr Roylance |