Public reporting of comparative information on healthcare quality of physicians and hospitals through ‘report cards’ is hailed as a plausible way to improve health care (Arrow, 1963; Akerlof, 1970; Stiglitz et al., 1989). Without this information, patients may choose their physicians based on more measurable characteristics (such as cost) or by word-of-mouth or other informal referral practices not obviously related to their needs.

There are two general types of healthcare report cards: those that measure outcomes and those that measure process. Reports of cardiac surgeons' and hospitals' risk-adjusted mortality rates following coronary artery bypass graft (CABG) surgery are examples of outcomes-based reporting (Pennsylvania Health Care Cost Containment Council, 1992; New York State Department of Health, 1993; California CABG Mortality Reporting Program, 2001; New Jersey Department of Health and Senior Services, 2003). Process-based report cards, often called quality indicators, report on rates of medical interventions, such as screening tests and medication usage, which are assumed to be related to outcomes. The Centers for Medicare and Medicaid Services (CMS) nursing homes report card, reporting on quality of care in nursing homes nationwide (2003), the Agency for Healthcare Research and Quality's congressionally mandated National Healthcare Quality Report, reporting on 150 measures of quality (2003), and the National Committee for Quality Assurance Health Plan Employer Data and Information Set (HEDIS), which includes quality indicators on health plan performance (2004) are examples of report cards that use process measures.

Risk and disclosure

In order to qualify as an instance of informed consent, a patient's decision to consent to an operation needs to be grounded on an adequate basis of relevant information. Without such a basis of relevant information, a patient's decision to consent to an operation is not an effective informed consent, and is not, therefore, sufficient to authorize that operation. Because many of the categories of information that inform effective decisions by patients to consent to an operation are categories of specialist medical information, patients must rely on the disclosure of such information by medical professionals. A patient must receive an adequate disclosure of a variety of categories of information that are relevant to their decision to undergo an operation, as a precondition to the provision of effective informed consent.

Exactly which categories of information should be disclosed, for the purposes of providing informed consent, is in dispute. What is not disputed by any commentators who accept a doctrine of informed consent is that a necessary component of disclosure, for the purposes of informed consent, is disclosure of the reasonably foreseeable risks of an operation. In this chapter, we argue that disclosures made for the purposes of obtaining patients' informed consent to an operation ought to include material information about a subcategory of risk information: that is, information about the ability of available surgeons to perform the operation in question.

Introduction

Report cards, also known as league tables, allow publication of outcome data that can reflect the performance of a particular hospital, clinical unit, or an individual doctor. Increasing interest is being focused on clinical report cards, particularly on their use in monitoring the performance of individual doctors and in making that data public. Report cards do have a number of possible roles including self-audit, accountability and to demonstrate safety and industry regulation. Almost 10 per cent of Australia's Gross National Product is devoted to healthcare, so it is an extremely important sector of government (www.aihw.gov.au). Other industries spending this level of tax payers' funds are accountable to the community and it certainly seems appropriate that healthcare is also kept under scrutiny. However, how far should that scrutiny extend and in what form should outcomes be made available to the community? Should the outcomes of individual doctors be made available or should this level of outcome data be retained within the craft groups for self-regulation and audit, leaving unit-based and hospital-based data available for publication and dissemination to the public. What other industries are subject to public distribution of the results of individuals within that industry?

In this chapter we explore the issue of report cards, specifically with regard to cardiac surgeons, who are currently at the forefront of this debate.

The public reporting of individual surgeon performance information encounters a variety of objections, from medical, economic and ethical perspectives. The most common ethical argument against publicizing surgeon-specific performance data is that the use of individual surgeon report cards leads surgeons to avoid operating on high-risk patients, because these patients are more likely to have unsuccessful outcomes, and such outcomes would have a negative impact on the surgeon's report card. In their discussion of report cards in the US, Green and Wintfeld (1995) write that:

The former President of the UK Society of Cardiothoracic Surgeons, Bruce Keogh made a similar comment in relation to the introduction of surgeon report cards in the UK:

And these comments are echoed by the President of the Royal Australasian College of Surgeons, Dr Russell Stitz:

The performance assurance mechanisms that have been proposed over the past decade or so have many aims, some laudable, some less so. Sometimes they are designed essentially as cost containment measures; sometimes they are motivated by a genuine concern for raising the quality of healthcare. In their paper, ‘Informed consent and surgeons’ performance', Clarke and Oakley (2004) argue that data on surgeons' performance should be collected and disseminated for another reason: to provide patients with information they need for genuinely informed consent. Clarke and Oakley suggest that promoting informed consent is vital, inasmuch as so doing respects patient autonomy; a good which is so significant that its promotion trumps most other considerations. Indeed, they give only one example of a good that is important enough to restrict (though not to violate) patient autonomy – surgeon's privacy with respect to their sexual orientation – and explicitly argue that even a reduction in overall surgical utility may not be a weighty enough consideration to justify a restriction on autonomy (2004, p. 19 and p. 23).

I suggest, however, that Clarke and Oakley are mistaken in thinking that respecting autonomy requires giving it weight sufficient to trump most rival goods. Respecting autonomy does not require maximizing it; it requires taking it seriously. We respect patient autonomy by always taking it into consideration in ethical decision-making, just as we respect a person by always taking her interests into consideration, not by treating her interests as trumping all rival goods (if it were the case that respecting a person required taking her interests as overriding, it would require a miraculous harmony of interests for us to be able to simultaneously respect many people).

Evidence-based patient choice

The concept of ‘evidence-based patient choice’ brings together two developments of ethical importance in contemporary medicine: evidence-based medicine and the growth of patient-centredness (Parker, 2001). The concept of evidence-based medicine, whilst problematic in many respects, encapsulates the belief that decision-making in medicine should be justified on the basis of good-quality evidence for the effectiveness of the intervention rather than on the basis of tradition, established models of practice, clinician preference and authority or other grounds. Patient-centred medicine too has arisen out of a concern with, and a critical response to, traditional medical practice and in particular to its over-emphasis on the authority of the health care professional. To some extent, this latter development has been driven by broader social changes outside medicine including a greater willingness to challenge the decisions of professionals including those of health professionals and to require such decisions to be both accountable and transparent. It is also related to relatively rapid developments in medical science and technology, which have created, along with social changes, an ever-increasing range of ethical questions with regard to which patient values vary significantly. Thirdly, and related to the other two, the move to patient centredness, and indeed to evidence-based medicine, has also been driven by increased media attention on developments in medical technology and by public and media discussion of scandals in medicine and in medical research.

The publication of outcomes information, or ‘report cards’, for individual surgeons can be argued for on three distinct grounds. One kind of argument appeals to healthcare quality, and focuses on the value of individual performance auditing for patient safety and for an evidence-based approach to best practice. A second kind of argument constructs the patient as a healthcare ‘consumer’ and appeals to a notion of consumer rights, such that patients have a right to comparative information about the healthcare products and services that they consume. Some proponents of this kind of argument believe that enabling patients to be more informed consumers will introduce productive market incentives into the healthcare system. A third kind of argument appeals to respect for patient autonomy and the requirement of informed consent to any medical intervention. I will refer to these arguments, respectively, as ‘the argument from quality’, ‘the argument from consumer sovereignty’ and ‘the argument from informed consent’. In advocating the publication of surgeon-specific outcomes data, it matters which argument we take to be fundamental, because the basic rationale for having surgeon-specific report cards has implications for the form, content and funding of such a system.

With respect to the argument from quality, the literature examining the effect of public reporting of comparative performance information on healthcare quality presents an increasingly compelling argument that such reporting is necessary for sustainable quality improvement (Chassin, 2002; Marshall et al., 2002).

Informed consent and surgeons' performance information

An important reason for providing patients with performance data on individual surgeons is to enable patients to make better decisions about surgery, as a part of the informed consent process. Surgeons' performance data can be utilized to enable a variety of types of decision that a patient may face. A patient can utilize performance data on individual surgeons to enable a choice between available surgeons. A patient can utilize surgeons' performance data when deciding between surgery involving an available surgeon and a non-surgical alternative form of treatment. Also, a patient can utilize surgeons' performance data to help decide whether or not to wait for a high-performing surgeon, who is not currently available, to become available.

Traditionally, performance data on individual surgeons have not been disclosed to patients, and such data have not usually been thought necessary to disclose for the purposes of providing effective informed consent. Canonical treatments of the doctrine of informed consent, such as Faden and Beauchamp (1986), do not consider the possibility of making such information available to patients. However, it has recently been argued that the doctrine of informed consent implicitly requires that surgeons' performance data be made available to patients (Clarke and Oakley, 2004). The gist of this argument is easy enough to grasp: it is uncontroversial that the significant and material risks associated with an operation should be disclosed to a patient who is contemplating that operation.

Introduction

There is little doubt that report cards provide valuable information to enable surgeons to improve outcomes for patients. However, we believe report cards for surgeons are only the first step towards a systematic improvement of all healthcare outcomes for all patients. There is no reason why report cards should be confined to surgeons, and our work with anaesthetists at Geelong Hospital confirms this view. Furthermore, we believe that the crude mortality rates, or even risk-adjusted mortality rates, currently provided to surgeons are the early primitive analyses that will be replaced by more meaningful performance charts as these become more accepted by the healthcare professions. Again, our work in Geelong supports this contention. We believe that local data collection by the healthcare worker is a reliable and trusted means of entering data into appropriate databases for analysis. Our reasons for reaching this conclusion relate to two considerations of the trust that this data requires. Firstly, the individual whose performance is to be assessed or monitored by the data collection must trust the data. Some of the early data collections were not trusted by the clinician participants, and this led to a mistrust of databases by clinicians in practice (Black, 1997, 1999a). However, the conclusions from the work of Professor Nick Black, an epidemiologist at the London School of Hygiene and Tropical Medicine, is that better databases will gain the confidence of clinicians; and collecting data relating to performance in a manner that clinicians believe contains the relevant data on their performance will improve the performance of individuals and organisations (Black and Payne, 2003).

Introduction

The monitoring of healthcare performance has a surprisingly long history, going back to the work of Florence Nightingale, who produced analyses of mortality outcome measures and campaigned for uniform hospital and surgical statistics (Spiegelhalter, 1999). More recently, there has been a renewed focus on monitoring clinical standards in many countries' health services, particularly in the UK in light of high-profile cases like Bristol (The Bristol Royal Infirmary Inquiry, 2001) and Shipman (Baker, 2001; Aylin et al., 2003a) and in the US with the Agency for Healthcare Research and Quality's Patient Safety Initiative (Agency for Healthcare Research and Quality, 2003) and the Institute for Healthcare Improvement's 100 000 Lives Campaign (Institute for Healthcare Improvement).

Monitoring of performance data is not straightforward, with many pitfalls and it is important to get it right. A recent report from the Royal Statistical Society noted that: ‘Performance monitoring done well is broadly productive for those concerned. Done badly, it can be very costly and not merely ineffective but harmful and indeed destructive’ (The Royal Statistical Society Working Party on Performance Monitoring of Public Services, 2003).

This chapter looks at:

• Sources of data, both administrative and clinical, and their quality

• Statistical issues around performance monitoring

• Methods and presentation of data

• How to deal with analyses suggesting poor performance

Sources of data

The Bristol Inquiry report concluded that ‘Bristol was awash with data’ (The Bristol Royal Infirmary Inquiry, 2001).

Ever since the Inquiry into paediatric mortality rates at the Royal Bristol Infirmary, there has been strong pressure in the UK for greater monitoring and reporting of the performance of surgeons. The inquiry, chaired by Sir Ian Kennedy, was asked not only to investigate the disproportionate number of deaths among infants operated upon in Bristol in the 1980s and early 1990s, but to make recommendations that might be acted upon throughout the National Health Service (NHS). Kennedy reported in 2002. Some of the recommendations went far beyond the question of how best to ensure the safety of paediatric surgery, or even of surgery in general. They called for a ‘patient-centred’ NHS. This theme was enthusiastically taken up in the government response to Kennedy, and there are noticeable affinities between the idea of patient-centredness and the idea of ‘patient choice’, which is at the heart of UK government health policy.

Patient-centredness in Kennedy's sense is broader than patient safety: it extends to public involvement in national policy-making, and public and patient involvement in decision-making structures of local NHS trusts. In my view, both of these ways of involving patients and the public are at best loosely connected to the problems in Bristol that prompted the inquiry. Nevertheless, they were accepted by the UK government in its response to Kennedy, and in subsequent policy documents. Partly as a result, Department of Health (DoH) policy now runs together, or comes close to running together, the answers to four distinct questions:

1. How can surgeons who are not equal to the kinds of operations they are attempting – whose patients avoidably die or suffer complications – be identified and retrained?

2. […]

Introduction

What ethical ramifications might the commercial context of much modern medicine have for the report card movement? We live in a world in which medicine in general is increasingly subject to market forces; not only are more and more goods and services commodified, and hence able to be procured on the open market, but within the public sphere, market-like accountability processes are increasingly set in place. We need to consider what implications this social context might have for the ethical status of report cards. Perhaps what is morally permissible in the context of public provision might transmogrify into the morally pernicious in a commercial environment. What difference, if any, might market forces make?

In examining this question, I shall assume that the market and market forces are here to stay and provide a background context for any public policy decisions in this area. My focus will be upon the moral legitimacy of report cards in a market context. In pursuing this agenda I identify three morally salient features of markets, that concern (1) the market as an information system, (2) the market as a distributive mechanism and (3) the market as an incentive system. I subsequently argue that close examination of these features provides genuine grounds for caution. The first worry involves the distributive consequences of report cards. In so far as report cards increase the levels of distributive inequality, this is a pro tanto reason against them. The second worry involves the transformation of our incentive structures.

The old song by Walter implies that, if we are trusted, we will feel the need to take up some personal obligation to those who trust us. This is interesting, as it suggests that trust is not merely a device for cementing some certainty and commitment in relationships, but imposes an obligation on the trustee that goes to the core of his character. If surgeons are to be trusted to look after the interests of their patients, does this mean that they can also be expected to develop a corresponding sense of duty that is an integral part of the way they see themselves? It would be good if this were so. This chapter examines the effect that trust, or lack of it, may have on surgeons and how it can influence the way they behave. The use of report cards may increase the knowledge we have about particular surgeons, but how does the revelation and application of that knowledge affect surgeons and their relationship with patients?

Difficulties interpreting information

In many situations in life we have to make decisions where we would like to have more or a deeper understanding of the information that is available. We may not be able to access what we need to know or this is couched in terms we do not understand.

Introduction

Since the Institute of Medicine (1999) published To Err is Human in 1999, many publications have discussed the need for different approaches to disclosing adverse events to patients, and the need to create a culture of safety within the healthcare system. Many of these articles begin with a clinician discussing an adverse event in which they were involved (Richards, 2000; Wu, 2001; Payne, 2002). Each individual story provides the medical and policy communities with an isolated view of an adverse event and the disclosure or non-disclosure of that event to the patient. There have also been research papers in the legal and medical literature that are designed to address specific areas of disclosure (Popp, 2003; Wu, 2000). Error disclosure is now required by ethicists, professional organizations and increasingly by regulatory bodies.

The goal of this chapter is to combine these accounts, stories and recommendations into a coherent roadmap for guidance in the field of disclosure. To accomplish this goal, we will begin by defining key terms, and will provide evidence that disclosure is a central part of fostering a safety culture. We will examine physician report cards and their relationship to disclosure policies. We will address the significant gap that exists between the principle of error disclosure and actual practice. Although most of the literature on disclosure is based on in-patient adverse event occurrences, most of healthcare occurs in the ambulatory setting.

Accountability

In the United Kingdom a key driver of reforms in healthcare has been the public perception that there has been a failure of regulatory mechanisms in healthcare. These concerns were particularly focused on the Bristol Royal Infirmary scandal. The ‘Kennedy Report’ from the subsequent national inquiry into this scandal concluded that the culture of medicine required systematic change. In particular, the Report recommended that the conformist ‘club culture’ of the British National Health Service be transformed into a ‘patient-centred’ culture in which the quality and safety of medical services becomes of paramount concern and the potential for substandard service to be provided and for attempts to hide evidence of substandard service is minimised (Bristol Royal Infirmary Inquiry, 2001). The broad thrust of the Kennedy Report was that this transformation is to be achieved by the use of mechanisms designed to ensure accountability. Although the Kennedy Report did not specifically recommend that surgeon-specific performance data be publicly released, it recommended (among other things) the publication of performance indicators that would enable evidence of underperformance to be identified early on, and would help improve healthcare safety and quality by discouraging underperformance in the medical system (Bristol Royal Infirmary Inquiry, 2001). The key recommendations of the Kennedy report were taken up enthusiastically by the Blair government, which has generally been in favour of establishing formal mechanisms to ensure accountability in the public sector (Smith, 2005).

What is the difference between biopsy and cytology?

A biopsy is the process by which tissue is obtained for microscopic or other investigation. Diagnosis of diseases based on examination of individual cells and small clusters of cells is called cytology or cytopathology. Biopsy may be diagnostic or therapeutic, whilst cytology tests may be used for diagnosis or for screening.

What are the advantages and disadvantages of these techniques?

A cytology specimen is usually easier to obtain, causes minimal discomfort to the patient, is less likely to result in serious complications, and is cheaper than a tissue biopsy. Although, in many clinical situations, the accuracy of cytology and biopsy is the same, in some situations a biopsy result is more accurate, e.g. in the diagnosis of a follicular thyroid carcinoma. The other advantages of a biopsy are that it allows assessment of tumour invasion and in some cases an excisional biopsy may be the only treatment required.

Describe the different forms of biopsy techniques

• Fine Needle Aspiration (FNA): the least invasive of the biopsy techniques and one of the first-line diagnostic procedures in the evaluation of a palpable breast mass. The procedure has the capability of being both therapeutic and diagnostic, e.g. during aspiration of a breast cyst the aspirate is often green tinged and serous (diagnostic), and aspiration should collapse the cystic cavity (therapeutic)

• Core needle biopsy (CNB): involves removal of a core of deep tissue usually using a Trucutt needle. Samples obtained with CNB are often large enough to provide detailed histological and architectural information including the type and grade of the tumour, its invasiveness, as well as hormone receptor status. This is an advantage over FNA, particularly with patients who have large masses suggestive of cancer

• […]