LAW & POLICY, Vol. 30, No. 1, January 2008 ISSN 0265–8240 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Blackwell Publishing Ltd Oxford, UK LAPOaw & Policy 0265-8240 © 2008 The Author Journal compilation © 2008 Baldy Center for Law and Social Policy XXX Original Articles MLAW & POLICY January 2008 cDonald WORKING TO DEATH Working to Death: The Regulation of Working Hours in Health Care FIONA McDONALD Recent research highlights significant risks associated with health professionals working long hours—risks to their health and safety, to the safety and quality of care provided to patients, and to public safety. This article undertakes a review of the various instruments used to regulate working hours in health systems, using six countries (Australia, Canada, Denmark, New Zealand, the United Kingdom, and the United States) and the European Union as primary comparators. The review demonstrates differences in the instruments used to regulate the issue in these countries and in the economic, social, and cultural factors that limit instrument choice and moderate instrument effectiveness. I. INTRODUCTION Recent research highlights the significant risks associated with health professionals working long hours—risks to the health and safety of individual health professionals, to the safety and quality of care provided to their patients, to the public, and to the system itself (see, e.g., British Medical Association 1998, 1999). The significance and salience of this issue has resulted in a plethora of activity by policy actors who use a variety of governance strategies to limit the hours worked by health professionals. The nature and context of the health system renders this task a complex one, particularly when compared with other professions whose hours of work are limited by regulatory schemes. There are also serious and significant risks associated with the regulatory schemes used to reduce working hours in the health sector. The most serious and significant of these risks is that, despite the best intentions of the regulators, regulatory schemes may actually worsen patient safety at greater cost to the health system. As such, how This research was supported by Health Canada, the Canadian Institutes for Health Research, the Alberta Heritage Foundation for Medical Research, and the Nova Scotia Health Research Foundation. Thanks to Christina Holmes, Dr. Jocelyn Downie, William Lahey, Dr. Mavis Jones, Bridget Lewis, the editors and the anonymous reviewers for their helpful comments. Address correspondence to Fiona McDonald, School of Law, Queensland University of Technology, GPO Box 2434, Brisbane QLD 4001 Australia. Telephone: + 61 7 3138 2010. E-mail: fiona.mcdonald@qut.edu.au. MCDonald WORKING TO DEATH 109 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy policy actors regulate long hours in the health system is an issue that warrants specific study and analysis, both for its contributions to our theoretical knowledge and understanding of regulatory regimes and for the future development of issue-specific regulation. To this end, this article reviews the instruments used by various policy actors to regulate working hours in health systems, using the configurations of instruments in six countries (Australia, Canada, Denmark, New Zealand, the United Kingdom, and the United States) and accordingly one supranational body, the European Union, as primary comparators.1 In the first section, I outline the problems with long working hours in the health sector and associated regulatory challenges. In the second section, I examine the instruments used to regulate long working hours and briefly discuss their strengths and weaknesses. However, policy actors do not employ these instruments singly. Accordingly, in the third section, I examine the configurations of regulatory instruments employed to address this issue in the countries under review and assess the lessons from their implementation, as well as any factors that limit instrument effectiveness. The review demonstrates differences in the instruments used to regulate the issue in these countries and in the economic, social, and cultural factors that limit instrument choice and moderate instrument effectiveness. In the fourth section, I examine how many hours and in what configuration physicians-in-training may work, as established by regulation, in each country (or supranational body). First, however, two comments. The governance approaches outlined in this article are ultimately very simple. Policy actors have framed the issue as follows: there are adverse consequences from health professionals working long hours, including negative effects on worker health and safety, the quality and safety of the health services provided to patients and on the safety of the general public. Therefore, a simple solution to improve health and safety for all is to reduce the hours worked by health professionals.2 Some professional groups, notably some nursing groups, contest this simple premise. In a nursing context, they argue it is not the number of hours worked that causes health and safety concerns for nurses and patients, but rather the demands placed on nurses within those hours. Put another way, nurses providing care to fewer patients, even if it is within a long working period, will best improve the well-being of nurses and patients. Thus, some nursing groups advocate nurse-to-patient ratios as a regulatory solution best able to improve safety (Buchan 2004). As the focus of this article is limitations on working hours for health professionals, nurse-to-patient ratios are not examined in this article, but such divisions in approach further demonstrate the complexities of regulating issues related to health and safety in health systems. Despite long working hours being an issue for many health professions, the analysis that follows focuses primarily on physicians, as concerns about physicians dominate discourse around this problem. The reasons for this are complex. The foundation of many modern health systems was a process 110 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy where the medical profession, hospitals, and the state accommodated each other’s needs in the public interest (Tuohy 1999), and therefore policy actors may be reflexively primed to address the needs of physicians above other health professional groups. Despite, or perhaps because of this status, physicians generally (although again not exclusively) are the locus of blame when things go wrong in the health system. The “malpractice crisis” throws into stark relief the needs of this group to mitigate litigation risks (Abraham 2002). Finally, physicians, especially those in training, have worked on average longer hours than other health professional groups for historically contingent reasons. II. THE WORKING HOURS PROBLEM Although in general the negative consequences of working long hours are well known, until relatively recently there has been little context-specific analysis of the impact of long working hours within health systems and for health professionals and patients. The impact of excessive working hours for health professionals includes negative effects on their health and on their relationships with others, including family, friends, and colleagues (British Medical Association 1998, 1999). The empirical evidence supports the conclusion that the long working hours of many health professionals are an occupational health issue, if not an occupational safety issue (British Medical Association 1998, 1999). It also supports the conclusion that a reduction in working hours results in an improvement in health outcomes for health professionals (Scott and Deloitte Ltd 1987; Stamp et al. 2005; Hutter et al. 2006). The long working hours of health professionals also emerge as a causative factor of preventable unsafe treatment and care. The Institute of Medicine (IOM) (2004), amongst others, confirmed the significance of the issue in respect of patient safety concluding that long work hours pose one of the most serious safety risks, because fatigue slows reaction time, decreases energy, diminishes attention to detail, and otherwise contributes to errors. The IOM estimates that the annual national costs of unsafe health services in the U.S. health system are a minimum of $17 billion and perhaps 98,000 lives (IOM 2000). There is little consensus as to whether regulation in this area is necessary as part of an occupational health and safety program, as one part of a broader initiative to improve the safety and quality of health services for patients, or whether regulation should address both concerns, an issue addressed in more detail later in this article. Placing limits on working hours also poses some more general regulatory challenges. For the purposes of analysis, I categorize these challenges as regulatory, professional, and resourcing issues. A. REGULATORY ISSUES Despite an apparent consensus around the need to reduce working hours, placing limits on working hours poses considerable regulatory challenges. MCDonald WORKING TO DEATH 111 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy The three regulatory challenges most closely considered in this research are: what are the appropriate time parameters within which health professionals should work; the effectiveness of regulation in improving occupational health and safety or patient safety; and which form, or forms, of regulation to use to address this issue. As to the first challenge, despite evidence that long working hours are problematic in terms of public safety and private well-being, researchers have not yet determined how many hours, or in what configuration, it is safe for health professionals to work. Policymakers also struggle with this issue, resulting in significant differences between the maximum hours health professionals are supposed to work between each country or supranational body, an issue discussed in more detail in the fourth section of this article and set out in Table 2 (see Section V). As to the second challenge, research tends to confirm that a reduction in working hours provides better health outcomes for health professionals (see, e.g., British Medical Association 1998, 1999). However, studies report conflicting outcomes of the impact of reducing working hours on patient safety. A 2004 meta-analysis noted that some studies concluded that errors reduced after working hours were limited, but other studies indicated no improvement or that error rates worsened (Fletcher et al. 2004). The relationship between managing health professional fatigue while ensuring continuity of care and the safety of patients is complex. It carries with it the risk that any initiative to reduce the working hours of health professionals, whether to improve occupational health or patient safety, may in fact compromise patient safety (Fischer 2004; Fletcher et al. 2004; Brennan and Zinner 2003; British Medical Association 1999). As to the last challenge, as discussed in this article, it is evident that there are a number of forms of regulation that regulators can and do use to address this issue. Differing motivations for regulation, that is, whether the regulation attempts to address the issue as an occupational health and safety issue or as a patient safety issue, may provide an explanation for differences in instrument choice and policy direction. This explanation is necessarily incomplete without an examination of the structural constraints within which regulators in this area operate. By structural constraints, I mean factors that relate to the economic, political, and cultural context that underpins regulation in each country or supranational body. Some of these I describe below, but this research indicates that structural constraints also include factors specific to each country, such as membership in supranational bodies or the design of the health care system in that country. B. PROFESSIONAL ISSUES Health professionals are not a homogenous group and any attempt to reduce working hours must take account of different professional needs. Professions have different work patterns, as do specialist groups within those professions. Professionals also may work in dissimilar work environments, for example, 112 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy acute care and primary care, and rural or urban care sites, with different demand and supply issues and different work patterns. These different needs also intersect with different professional cultures associated with the various health care professions. By culture, I mean the values that define professionalism in a particular profession. Changing a profession with an entrenched culture through regulatory means may be especially challenging, as regulatory strategies may fail if they overlook the subtleties of the behaviors and cultures of the regulated (Aiken and Sloane 2002). Many professions cling strongly to their professional autonomy and are resistant to external regulation of their practice (Eastman and Peay 1999). This culture includes a tradition of working long hours, most particularly those who work or train in the acute health sector. The medical profession has traditionally perceived the ability and willingness of a physician to work long hours as demonstrating commitment to the profession and to patients (Fischer 2004; Carpenter et al. 2006). In particular, society expects health professionals to be available when patients are in need; limits on working hours may place that professional tradition in question (Carpenter et al. 2006). Conversely, health professionals also commit to the ethical principle of “first do no harm” (Sharpe 2004), and working long hours may cause harm, placing many health professionals in an ethically conflicted position (Carpenter et al. 2006). The professions, particularly the medical profession, are also concerned about ensuring the quality of the training provided to health professionals. This concern is particularly prevalent in the United States, where a number of critics, including some students, are concerned that limits on working hours (in the United States approximately eighty hours per week) negatively affect physician training (Green 1995; Hutter et al. 2006; Brunworth 2006; Cohen-Gadol et al. 2005; Barden 2002; Sakorafas 2004). In contrast, in the United Kingdom, the Royal Colleges accept that medical training can occur satisfactorily within fifty-six hours per week (Pickersgill 2001; Department of Health 2002), and in Canada an eighty-hour work week is considered sufficient to meet training needs for physicians (Romanchuk 2004). Putting cultural factors and training issues together, there may be substantial pressure in some health systems to resist governance strategies to reduce working hours. C. RESOURCE ISSUES Another significant challenge facing would-be regulators are concerns about the continued sustainability of health systems—systems confronted by increasing demand and limited resources, fiscal and human. A decision to reduce working hours has an inevitable impact upon resources, as health systems will have to employ more health professionals to cover the shortfall in hours. There will be an inevitable fiscal impact, as costs will increase, at least in the short term, unless and until they are offset by savings from MCDonald WORKING TO DEATH 113 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy fewer incidents of unsafe care and lesser rates of illness, burnout, and stress amongst health professionals. In environments where fiscal resources are constrained, decision makers face a dilemma as to whether to commit to significant up-front costs, which offer only the possibility, and not the certainty, of future benefit. There are also questions about workforce availability, as there is an international shortage of health professionals, a shortage exacerbated if health professionals are required to work shorter hours (U.K. Hansard 2003). There are also questions about the willingness of the workforce to work reduced hours. Working fewer hours may mean health professionals receive less pay or allowances, which would be of concern for health professionals whose salaries tend toward the lower end of the scale, for example, nurses, and for newly qualified health professionals with significant debt levels. Fiscal and workforce issues are not analyzed in any detail in this article, but are significant issues that affect the manageability and effectiveness of any governance regime that intends to limit health professionals working hours. III. OVERVIEW OF AVAILABLE REGULATORY INSTRUMENTS Governance within the health system has long reflected the central reality of public problem solving, namely its collaborative nature (Braithwaite, Healy and Dwan 2005) due to its accommodatory base (Tuohy 1999). Governments lack expertise in the delivery of health services and so must accord considerable autonomy to health professionals, with problem solving and policymaking within the health system generally relying on government and third-party actors. This has generally occurred through coregulatory models, but also through self-regulatory interventions supported by government (Braithwaite, Healy and Dwan 2005). Government use of command and control regulation has generally only occurred where the market, self-regulation or coregulation are perceived as having failed to serve the public interest. The six countries (and one supranational body) examined in this article generally employ a mix of regulatory instruments to regulate working hours. The basic instruments identified in this review are set out in Table 1, and the strengths and weaknesses of these instruments are subsequently discussed in more detail. I consider that a regulatory instrument is strong if it is: effective, efficient, equitable, manageable, and politically legitimate (Salamon 2002; Gunningham and Grabosky 1998). I define “effective” in simple and complex terms. The simple question is whether regulation reduces the working hours of health professionals; the more complex question is whether regulation contributes to improving the health and safety of patients and/or health professionals. Adapting Salamon’s (2002) definitions, an instrument is efficient if it reduces working hours and/or improves safety at minimum cost in an administratively efficient manner. It is manageable if there is relative ease in its operation (Salamon 2002). An instrument is equitable if 114 LAW & POLICY January 2008 Journal compilation © 2008 Baldy Center for Law and Social Policy © 2008 The Authors Table 1. Basic Instruments Employed to Regulate Working Hours in Health Systems Instrument Type Policy Actor(s) Strengths Weaknesses Tort Law—medical malpractice/general negligence (regulation by litigation) Patients/Public at large Effective for some affected individuals (compensation) Ineffective (simple & complex) at systems level Inefficient Inequitable Questionable political legitimacy Tort Law—employment (regulation by litigation) Employees Effective for affected individuals (compensation) (working conditions) Ineffective (simple & complex) at systems level Inefficient Inequitable Questionable political legitimacy Labor Law—collective agreements (co-regulation) Unions/Employers Gradual implementation Mutual agreement Equitable (if national agreement) Efficient Simple effectiveness Complex effectiveness—improvements in worker health Manageable Possible inequities Questions of legitimacy Complex effectiveness— patient safety improvement uncertain Self-Regulation Professional bodies/ accreditation agencies Simple effectiveness Complex effectiveness—improvements in worker health Equity within group Questions of legitimacy Complex effectiveness—patient safety improvement uncertain Efficient? Manageable? Command and Control Regulation Government Equity Simple effectiveness Complex effectiveness—improvements in worker health Politically legitimate Complex effectiveness—patient safety improvement uncertain Imposed solution Efficient? Manageable? MCDonald WORKING TO DEATH 115 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy it shows fairness in burden sharing between parties (Salamon 2002). An instrument is politically legitimate if it supports autonomy, transparency, and accountability (Salamon 2002). A. REGULATION THROUGH LITIGATION The adverse consequences of long working hours fall heavily on individuals. Some argue that it is not necessary to place limits on working hours in the health system, as patients can sue for errors attributable to excessive hours worked by health professionals and the employment practices that permit them (Annas 2006).3 For example, in Australia, the judge in Brotherston v Royal Perth Hospital (1995) awarded substantial damages to a man whose permanent brain damage was a result of negligent medical treatment. The judge concluded that it was difficult to avoid the conclusion that the long shifts worked by staff, with an overwhelming workload, contributed to poor clinical management. In England, the court in Wilsher v Essex Health Authority (1986) awarded damages to a baby whose physicians, who had been working continuously or had averaged long hours on call in the prior weeks, repeatedly inserted a catheter for arterial gas sampling into a vein, rather than an artery. The baby became blind after thirty-two hours of supersaturation with oxygen. The judge held the hospital liable for its failure to organize its staff appropriately. The public may also sue for acts of negligence that occur outside the health system, but that may be attributable to its management practices, for example, fatigued health professionals’ high rates of vehicle accidents (Barger et al. 2005). In the United States, for example, a physician, and the hospital the physician worked at, were sued after the physician rear-ended a woman’s car and seriously and permanently injured her. The physician was returning home after working a thirty-six-hour shift (Gotbaum 2005). While the case proceeded against the physician, the court of first instance dismissed the action against the hospital stating that it did not owe a duty to a third party injured by an off-duty physician who allegedly experienced fatigue due to the hospital’s policy on working hours (Brewster v RushPresbyterian-St Luke’s Medical Center 2005). This decision was upheld on appeal.4 Employees can sue employers for loss or injury caused by the adverse effects on health and safety of their conditions of work (Patton 2001). In the English case of Johnstone v Bloomsbury Health Authority (1991), for example, Dr Johnstone claimed that he suffered a personal injury as a result of his work schedule (including one shift of forty-nine hours, with seven hours sleep, and another of thirty-two hours with thirty minutes sleep). The Court of Appeal found that a Health Authority could not make a physician work so much overtime that it constituted a threat to his or her health. Lord-Justice Stuart-Smith stated it was “a matter of grave public concern that junior doctors should be required to work such long hours 116 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy without proper rest that not only their own health may be put at risk but that of their patients as well” (ibid.: 301). Tort law serves two regulatory functions: social insurance by compensation for individuals harmed by others (a reactive function); and deterrence, where the award of financial compensation creates economic incentives for defendants to implement risk management systems to ensure safer conduct in the future (a prospective function) (Sharpe 2004; Abraham 2002). Due to its prospective function, simple tort litigation can claim to be an instrument for regulation. For most countries, until very recently, tort litigation has been the primary instrument for regulating patient safety (Mello, Kelly and Brennan 2005), although not necessarily worker health and safety where key regulatory instruments are employment agreements and, to a lesser extent, occupational health and safety laws. From the case law discussed above, it appears that the impact of regulation by litigation upon employment practices within the health system is limited in its effectiveness. An individual employee may directly challenge his or her employment conditions and effect change, at least for that health professional. Patients may argue working hour related fatigue as a systemic factor establishing liability of hospitals when there is evidence to support it to claim compensation, whether that results in systemic change is open to question. However, the limited U.S. case law in this area indicates that the general public will not be able to use litigation in respect of motor vehicle accidents to compel changes in employment practices in health sectors. This is despite empirical evidence that long working hours are causally connected to higher than average rates of motor vehicle accidents experienced by physicians returning home after work (Barger et al. 2005). Recent U.S. experience points even more strongly to litigation as an explicit regulatory instrument, which has as its goal promotion of prospective systemic change. In the United States, litigation may result in negotiated regulatory policies to settle suits or provide a lever to promote support for government action (Viscusi 2002). Characteristically, areas that result in what Viscusi (ibid.) terms “regulation through litigation” are harms which generate significant numbers of high-profile individual or class actions, for example, tobacco, breast implants and firearms (ibid.). Most countries, at least those with fault-based systems for medical malpractice, are experiencing a high-profile “malpractice crisis” (Annas 2006; Mello, Kelly and Brennan 2005). The outcome of the malpractice “crisis” and a general awareness of patient safety costs for regulators generally is a renewed interest in encouraging effective self-regulation, a move to metaregulation, and the institution of command and control models of regulation where required (Downie et al. 2006). Although state governments in the United States have aligned themselves to support individual suits as levers for policymaking or instituted litigation as a form of associational self-regulation (Viscusi 2002), this has not yet happened in respect of the regulation of working hours in health systems and is generally not a use of litigation contemplated in the countries examined in this article. MCDonald WORKING TO DEATH 117 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Tort law has notable weaknesses as a regulatory instrument to promote systemic change. It is bottom-up in nature, as it depends upon the willingness and ability of individuals to seek a remedy through the sometimes torturous processes involved in taking a matter before a court or administrative tribunal. There are barriers to many individuals bringing cases before the courts, including informational and financial barriers, and for employees, the possibility of impaired career prospects. In the absence of action by individuals, courts cannot provide guidance on the expected standard of care. Even if an individual takes action, the matter may settle before the matter reaches the courtroom, as these are generally private disputes. With no transparency as to outcome, there can be no broader systemic change. There are a number of additional barriers to litigation achieving even simple effectiveness and achieving complex effectiveness is even more fraught with problems. One of the central arguments in support of regulation through litigation is that large damages awards provide a strong incentive for administrators to limit working hours on a cost/benefit basis (Patton, Landers and Agarwhal 2001). This effect is strongly contested (Danzon 1985; Mello and Brennan 2002), and any assumption of a broader effect of litigation on the policies and practices of other health providers is even more strongly contested (Mello and Brennan 2002; Mello, Kelly and Brennan 2005). In part, this is because all health providers, individual and institutional, carry insurance and in part because of a sense of defensiveness about lawsuits, by physicians in particular (Mello, Kelly and Brennan 2005). The uneven impact of litigation will produce inequity of expected working hours across the sector with the result that health professionals in different facilities will work in widely varying conditions, and patients will get varying treatment and care that may be more or less safe. Efficiency is also in question, with court processes resulting in slow, diffuse regulatory change, instead of coordinated systemic changes to enhance safety across the system (Jacobson and Soliman 2002). It is also unwieldy, with multiple lawsuits being an inefficient manner to provoke systemic change. Lastly, the tort system raises questions about political legitimacy. If sectoral change is possible through tort litigation, do we want the courts to change employment practices in the sector, or should another policy actor take responsibility for change? Courts often act in a policy capacity only when those in a position to enact a regulatory change appear to ignore an issue of significant public interest, such as safety (Viscusi 2002; Mello, Kelly and Brennan 2005). Setting aside for the moment that courts act in a reactive manner, courts may lack expertise when it comes to some policy issues and provide a remedy that is impracticable to implement, does not take into account the impact of the judgment on other important public policy issues, such as health human resources, or could have the unintended effect of worsening patient safety. Lastly, there is also a lack of transparency in this type of regulation, which essentially is a private negotiation to which the public are not privy and which may not serve the public interest. Although 118 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy the ability of individuals to bring a suit against an institution whose employment practices result in harm is available, in some form or another, in all the countries examined in this article, it has, to date, not resulted in significant changes in employment practices across systems. B. CO-REGULATION In a number of settings collective associations of employees (unions) have acted, in negotiation with employers, to change employment conditions including reducing working hours. All of the countries examined in this review are able to use, or have used in the past, this mechanism to regulate employment conditions in health systems. Collective negotiations fit loosely within a model of coregulation, where policy actors cooperate to create new rules for a specific context. Such negotiated policies pursue public and private interests. For the majority of the countries examined in this research, collective agreements were a viable regulatory instrument; only the United States is an outlier. Collective agreements were most successful in countries with a tradition of government involvement in the universal provision of health services. In these countries, legislative frameworks centralized publicly funded health services within structural frameworks at the local, regional, or national level. Further, the regulatory traditions within the health sectors of these countries were based on accommodations with professional groups, in particular physicians (Tuohy 1999). Therefore, the negotiation of collective agreements at a national or subnational level between health professionals and government or government actors was an accepted part of the governance framework. In these circumstances, unions of health professionals often had considerable power in negotiations. This power was even more compelling when policymakers were confronted by the monopoly these unionized professional groups had in the provision of an essential service. In contrast, in the United States, governments play a lesser role in the provision and funding of health services, with many health services provided privately by a multiplicity of providers and many health services directly funded by private health insurance plans. The presence and influence of unions in the U.S. health sector is less overt and powerful, especially in regard to physicians, than in the other countries examined in this research. One barrier to the effectiveness of negotiated agreements is how well parties to that agreement comply with it. Generally compliance with negotiated agreements is high, for example, in Denmark and New Zealand, the working hours of the vast majority of physicians affected by collective agreements decreased in line with the limits specified in their collective agreements, although the occasional egregious breach was reported (Child and Old 2004; Australian Medical Association 1998). The United Kingdom experienced greater levels of noncompliance with its negotiated agreement. A 1997 examination of compliance indicated that working hours had reduced for MCDonald WORKING TO DEATH 119 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy the majority of physicians-in-training, but 21 percent of posts did not comply with the targets for actual hours of work set out in the collective agreement (Australian Medical Association 1998). There is limited evidence of effectiveness in terms of improving patient safety and the quality of health services and of worker health and safety. In New Zealand, a 1987 independent review for the Department of Health of current medical officer rostering practices assessed the impact of a collective agreement that reduced physicians-in-trainings’ hours. The review concluded that residents were less tired and better able to cope than before limits were in place and that there was an improvement in some aspects of patient care (Smakoff and Jacques 1991). In terms of political legitimacy, coregulatory approaches to regulation may give rise to concerns about conflicts of interest and regulatory capture, as such decision-making processes endeavor to balance public and private interests (Gunningham and Grabosky 1998). This form of negotiation may privilege certain actors to address issues that are of public interest and exclude others from knowledge of or participation in negotiations. Negotiations and agreements may not be transparent as they may not be available, accessible, or made known to the public, as in New Zealand where the terms of the collective agreement only became public currency in the context of a bitter disagreement about conditions. There may also be questions about the accountability of health professionals, institutions, or government to the public for decisions that may have an impact on the safety or the quality of the care patients receive. Collective agreements are, to a degree, self-policing and therefore reduce enforcement costs, making them, at least in this respect, a potentially costeffective mechanism, especially when compared to command and control mechanisms. In the United States, effective implementation of the command and control model used in New York State and the associational selfregulatory model implemented by the Accreditation Council for Graduate Medical Education (ACGME) in all other states required some investment in compliance mechanisms, such as the employment of inspectors. In contrast, the coregulatory model used in New Zealand required no financial outlay in respect of ensuring compliance. The level to which these mechanisms can be effectively self-policing does depend to an extent on individuals informing unions of violations, and there may be significant barriers to this occurring, either internal or external in nature, as is discussed in more detail later in this article in respect of ACGME’s experience (Croasdale 2004). Collective agreements generally come into force in gradual increments that enable affected institutions or professions to adjust practices and adapt to change more slowly, while working toward an ultimate substantive reduction in hours. This approach may reduce resistance to change. Sometimes the pace of change can be too slow. For example, in New Zealand physicians-in-training have relatively recently taken strike action in respect of working hours claiming progress has stopped (New Zealand Press 120 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Association 2006). However, it generally seems that for some groups, in some countries, coregulatory instruments have brought about real change in employment practices. C. SELF-REGULATION Health system management internationally is heavily reliant on health professionals and health providers regulating themselves (Mello, Kelly and Brennan 2005). Faith in the ability of health professionals to self-regulate in part relies on the professions’ adherence to their professional codes, which require health professionals to act in the interests of patients. Selfregulation is also premised upon the idea that professions (or institutions) have the specialized knowledge necessary to determine how best to provide clinical services. There are three primary self-regulatory actors in this sphere: regulatory colleges/councils (bodies sanctioned by government to govern the professions); professional associations (interest groups representing the interests of members); and accreditation bodies. There are various models describing hierarchies in regulation but for the purposes of this article I use the model of ascending hierarchies of coercion (Braithwaite, Healy and Dwan 2005; Gunningham and Grabosky 1998). The hierarchy begins with bottom-up regulatory approaches that rely on individuals or organizations to voluntarily self-regulate to change culture and behavior at individual or systems levels. This is relatively uncommon, although at least one professional association is seeking to extend its regulatory reach through creating voluntary guidelines about working hours and working environments (Australian Medical Association 1999). Using guidelines as a mechanism to enable systemic change tends to be ineffective as compliance is voluntary and behavior and practices are difficult to change. Institutional accreditation bodies, such as the Joint Commission in the United States, do not appear to directly address limitations on working hours; rather they may address the organizational impacts of a reduction in working hours indirectly, for example through handover protocols (Okie 2007). The next step on the hierarchy is associational self-regulation, where a policy actor places pressure on another actor, usually through intimating it might legislate or instituting lawsuits, to encourage/compel “voluntary” selfregulation (Viscusi 2002). The United States arguably has employed this model, as critics suggest that government pressure, whether real or imagined, influenced the ACGME’s5 2003 decision to institute mandatory standards to regulate the working hours of physicians-in-training (Bell 2003; Altman 2002; ACGME 2003). While physician groups may perceive coregulatory processes as acknowledging their professional autonomy, they may perceive associational self-regulation as undermining it. It may be perceived as the imposition of a regulatory framework that may not take into account professional values or the resource constraints within which professionals must operate (Kagan and Scholz 1984). This may explain why MCDonald WORKING TO DEATH 121 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy many U.S. physicians strongly oppose the ACGME’s standards. Associative self-regulation may therefore be more likely to be ineffective both in simple and complex terms. Although such regulation is likely to be equitable, as standards generally have national scope, it is also likely to be inefficient because of high compliance costs. The third level in the hierarchy is state-sanctioned self-regulation, where the state delegates coercive powers of top-down regulation to the profession(s) to self-govern. Empowering legislation generally requires regulatory councils to address specific issues related to the functioning of the profession. These include registration, matters of professional practice and conduct, and, at times, the development of ethical standards for the profession. The expressed purpose of most regulatory councils in legislation is protection of the public (Downie et al. 2006). Despite this clear statement of purpose, and evidence of the impact of long working hours on the safety of the public, regulatory councils/colleges in the countries reviewed appear not to have created any policies about this issue or indeed to have acknowledged the issue. At the microlevel, disciplinary tribunals of the regulatory councils/ colleges address individual cases where fatigue is alleged to have contributed to misconduct, with whatever macro-level deterrent effect that this might have. For example, in Medical Board of Queensland v Doneman (2004) the Tribunal held that the requirement that the physician work long hours must have contributed to his ability to “properly manage” the patient’s condition. It concluded “if this tragedy leads to nothing else, it should lead to the abolition of such brutally long shift hours, which must in itself reduce the standard of care available to patients” (ibid.: para. 6). There are also concerns about the legitimacy of self-regulation, as it may raise concerns about conflicts of interest (perceived institutional reluctance to monitor peers and enforce sanctions), public accountability and about the transparency of processes (Ogus 1995; Anonymous 2002). Last in the hierarchy is “metaregulation” where government agencies require and oversee self-regulatory action (Grabosky 1995; Gunningham and Grabosky 1998). Despite a governance trend internationally away from professional, or institutional, self-regulation of issues pertaining to patient safety toward metaregulation (Downie et al. 2006), governments, at least in the countries reviewed for this article, have not instigated any form of metaregulation in respect of working hours. Metaregulation encompasses meta-risk-management schemes, enforced quality improvement, or enforced self-regulation (Braithwaite, Healy and Dwan 2005). Potential advantages of this approach are that it accords sufficient flexibility to address local circumstances while at the same time setting a minimum standard to ensure safety. It is also transparent and ensures accountability. Metaregulatory approaches to limiting working hours are currently in place in some, nonhealth related, sectors with varying accounts of their success (Downie et al. 2006). Aside from collective efforts in respect of employment conditions, professional self-regulation of long working hours has been somewhat limited. It 122 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy may be that self-regulatory bodies are inactive because of conflict within the profession about the necessity, or desirability, of limiting working hours. Inaction may also be a tacit acknowledgement that other actors are better equipped to address this issue, or other policy actors have, however, imperfectly, already addressed it. D. COMMAND AND CONTROL REGULATION One of governments’ responsibilities is to mitigate risks to the public health and safety where possible. This does not mean government must always directly intervene; if there are existing successful governance mechanisms put in place by another policy actor, no action may be necessary. A failure by other policy actors to take action is a problem for government. Indeed, the failure of the profession to act appropriately was one expressed rationale for U.S. federal legislators introducing legislation to limit working hours (Patient and Physician Safety and Protection Act 2005). As Walshe notes, the problem with relying on voluntarist types of regulation is that actors may be constrained from implementing internal change due to very real environmental and cultural issues (2003), whereas command and control regulation instructs the regulated to operate within those constraints. Command and control regulation can rely on voluntary compliance of the regulated if regulators consider them to be “responsible political actors,” to use terminology from Kagan and Scholz’s typology (1984), or “virtuous” actors, to use Braithwaite’s description (2002), and so likely to comply. Given the ethical commitments of health providers to “first do no harm” (Sharpe 2004) and that violators would pay a heavy price in any litigation establishing fatigue as a causative factor in error, regulators could perhaps assume responsibility and virtuosity. However, there are other countervailing factors such as culture, resourcing, and autonomy. Kagan and Scholz (1984) suggest that noncompliance results when the regulated consider regulation unreasonable. The tendency of health professionals and health administrators to ignore or manipulate laws they do not agree with or cannot comply with is well-known (Eastman and Peay 1999; Walshe 2003). Thus, legislation of this type may require investment in compliance measures to ensure simple effectiveness, increasing costs and adding to manageability issues. Generally, command and control regulation is considered an effective mechanism, both in terms of achieving desired safety outcomes and as a tool to ensure consistency across a sector or across multiple jurisdictions. Gunningham and Johnstone (1999) and Gunningham and Grabosky (1998) note that command and control regulation remains an effective instrument to improve health and safety in some contexts. The question must remain whether it will be equally effective in this context given the complexity of health systems. In contrast with other systems, in health systems health and safety risks are not just to employees but also to patients. Regulators may also see command and control MCDonald WORKING TO DEATH 123 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy regulation as a better option for when they want, as in the example of the European Union, to implement cross-sectoral change, as it ensures consistency of approaches between sectors and between countries and equity of treatment. One of the important functions of legislation is its symbolic effect in sending a signal that some employment practices create unacceptable health and safety risks (Gunningham and Grabosky 1998). Legislation sets clear expectations about working hours and enables accountability both of the regulated for performance and the regulators for a lack of monitoring and enforcement (Gunningham and Grabosky 1998). In New York State, revelations that most facilities were ignoring legislated limits on working hours compelled substantive change to the practice of regulators to ensure that the regulated were publicly accountable for their lack of compliance. A legislative scheme may be transparent, especially if regulators publicly report compliance, as is the case in New York State. To the extent a legislative schema depends upon reporting of abuses by health professionals, it is vulnerable to criticisms, discussed subsequently that are leveled at any reporting scheme (Croasdale 2004). Perhaps the strongest criticism is that law is a blunt instrument incapable of adequately addressing complex issues relating to the management of individuals who work highly complex, highly integrated jobs, driven by unexpected emergent human needs. Critics argue that the nature of patient care means that sudden emergencies may arise or treatment may prove more complex than initially expected, and the patient’s life may be at risk if physicians are less attentive because their shift is about to end (Steinbrook 2002). An examination of actual regulatory instruments within and outside of health systems in relation to working hours confirms that legislation need not be inflexible, as some legislation imports qualifying words allowing overrides in emergency situations (Downie et al. 2006). Regulation can also be crafted so that enforcement strategies, at least initially, fall within the compliance (education and negotiation), rather than deterrence (punishment), model of enforcement (Day and Klein 1987). Whether and when inspectors move from compliance to deterrence strategies will depend on the interpretative judgment of the regulator as to the behavior of the regulated, that is, whether the regulated are deliberately trying to subvert the system or whether they are genuinely constrained by the environment in which they function (Kagan and Scholz 1984; Day and Klein 1987). There is precedent for the use of command and control regulation in complex areas and its relative effectiveness, and a place for law when other policy actors fail to act to preserve the public interest (Gunningham and Grabosky 1998). IV. CURRENT REGULATORY CONFIGURATIONS An examination of the process of regulation demonstrates that once policy actors accepted that long working hours were a problem, whether because 124 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy of concerns about the welfare of workers or the safety of patients, no country has employed a single instrument model. There are three general macrolevel approaches to this issue displayed in the configurations of regulation demonstrated in each country, or transnational block, and these are illustrated in Figure 1 in order of the degree of coercion involved. A. UNITED STATES (EXCEPT NEW YORK STATE) In the United States, until very recently, the only mechanism through which to regulate the working hours of health professionals was when individuals used tort law, except in New York State. If changes to employment practices have resulted from litigation, they remain facility specific. For historically contingent reasons, collective mechanisms were not available to physicians-in-training until 1999. In Boston Medical Center Corporation v House Officers’ Association/Committee of Interns and Residents (1999), the National Labor Board overturned twenty years of precedent and recognized that physicians-in-training were employees, allowing them the right to collective bargaining. Despite this, collective negotiations appear infrequent, with the Committee of Interns and Residents, the largest medical resident union in the United States, representing only 12 percent of all physiciansin-training (Butterfield 2007). The primary mode of regulation in the United States is associational selfregulation through the Accreditation Council for Graduate Medical Education (ACGME). ACGME must accredit training facilities in order for those facilities to gain government funding. In 2003, ACGME introduced Resident Duty Hours Standards as part of its criteria for accreditation.6 Critics noted that the introduction of the standards coincided with proposed federal legislation to limit working hours for physicians-in-training7 because of patient safety concerns and a petition by physicians-in-training asking the federal Occupational Health and Safety Administration to intervene, which it declined to do (Bell 2003; Altman and Grady 2002). ACGME said that it regulated working hours as training facilities had implemented service cuts, because of shorter hospital stays, technological advances, and financial pressures, which resulted in increased pressure for physicians-in-training to Figure 1. Configurations of Regulatory Approaches to Limiting Working Hours. Configuration one Configuration two Configuration three United States Australia, Canada, N.Z. European Union/New York State (Except NY State) Regulation by litigation Regulation by litigation Regulation by litigation Self-regulation Limited self-regulation Limited self-regulation Coregulation Legislation MCDonald WORKING TO DEATH 125 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy work longer hours, thus jeopardizing the quality of care provided to patients and the standard of education provided to physicians-in-training (ACGME 2003). It is important to note that ACGME’s decision occurred in a policy climate where the IOM’s report on the scope and nature of the patient safety problem in U.S. health systems had received national attention and become a national priority (1998). It was patient safety rather than any overwhelming anxiety about occupational safety and health that prompted regulation in the United States. The ACGME scheme is of particular interest, as it is a self-regulatory scheme that is mandatory in nature but instituted as the result of (real or perceived) pressure from government. ACGME faced much opposition to its introduction, and it remains controversial, with critics complaining about its effect on the quality of medical education and claiming negative impacts on patient care (see, e.g., Cohen-Gadol et al. 2005; Brunworth and Sindawi 2006; Carpenter et al. 2006). The evidence of simple effectiveness in terms of reduction in working hours is mixed. Although many programs are broadly compliant with the new standards, some few remain noncompliant (see, e.g., ACGME 2006). ACGME conducts prearranged site visits every three years and reports it is seeing evidence that some logbooks recording work hours are being “fudged” indicating a degree of resistance to the standards (Croasdale 2004). Physicians-in-training report that a significant number feel compelled, primarily, they state, out of ethical concern for patients, to exceed the ACGME standards and to falsify records to obscure this (Carpenter et al. 2006). In terms of complex effectiveness, there are also reservations about whether the ACGME governance regime is effective. A recent examination of error rates in obstetrics and gynecology before and after implementation of the ACGME guidelines concludes that physician-in-training work hour restrictions show minimal evidence of improvement in quality of care (Bailit and Blanchard 2004). Physicians-in-training and the physicians who supervise them also report perceptions that the quality and safety of the services provided to patients have diminished after the introduction of the ACGME standards (see, e.g., Carpenter et al. 2006; Cohen-Gadol et al. 2005; Brunworth 2006). Even assuming hours have reduced, there is still a question as to whether the reduction is sufficient to address issues of employee health with some critics concerned that the maximum hours are still too long (Parshurum et al. 2004; American Medical Student Association 2005). In terms of both equity and manageability concerns remain. The ACGME standards have an opt-out clause for some programs for “educational reasons.” Exploiting this, some specialties may be able to work still longer hours resulting in inequitable treatment of physicians-in-training, depending on their specialty (i.e., surgical trainees versus medical) or the hospital or region that they work in. The manageability of the regulatory schema is also in question since ACGME relies on physicians-in-training to report violations of the standards. A significant issue not addressed by 126 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy ACGME, and which is the bane of all reporting programs, is how to overcome systemic and cultural barriers that actively discourage reporting. Physiciansin-training may face repercussions for reporting, as ACGME does not provide whistleblower protections (Hexom 2004). The only penalty the ACGME can impose against violators is loss of accreditation. While loss of accreditation is highly stigmatizing for an institution, it may also negatively impact upon the careers and prospects of those in training in that institution discouraging reporting (Croasdale 2004). Loss of accreditation is also such a serious penalty that there may be some hesitation on the part of the ACGME to apply it. The ACGME has had some form of work hour restrictions in place since 1987, but, despite a significant number of citations (ACGME 2006),8 only one program lost accreditation.9 Despite promising to aggressively monitor compliance with the standards, there remain concerns about the conflict of interest faced by ACGME (Bell 2003; Anonymous 2002). A critic noted that “Despite tough talk, the council faces an inherent conflict of interest. Its board is dominated by the trade associations for hospitals, doctors and medical schools, all of which benefit from the cheap labor provided by medical residents” (Anonymous 2002: A36). In addition, critics argue that “the use of public dollars to fund graduate medical education should require public accountability” (Boodman 2001: H01). ACGME does not make public statements about why it is withdrawing accreditation or placing a program on probation, so there is limited accountability to the public that funds the training programs (Hexom 2004) and little transparency. Self-regulation can be an important instrument for regulatory change. However, in this case, when the regulated perceived regulation as being imposed upon the profession because of government pressure, there was significant opposition to the standards. This opposition undermined regulatory effectiveness. Opposition did not only relate to the manner in which working hours were regulated but also the fact that it was felt necessary to regulate the issue at all. As noted earlier, in the United States, the culture of the medical profession was generally one where working long hours is considered essential to good medical practice and to effective professional training. Elements of the profession were unhappy that regulation altered these, as they saw it, core professional values. Additionally, the pressure to regulate implied that the profession was not rigorous in upholding its core ethical value “first do no harm” and thus was not a good ethical actor. These factors point to the need to have at least some support from key policy actors within the health professions before implementing regulatory change. B. AUSTRALIA, CANADA, NEW ZEALAND Regulation through litigation is an instrument used to regulate working hours in Australia and Canada. It is much more limited in its scope in New MCDonald WORKING TO DEATH 127 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Zealand, as New Zealand’s no-fault system precludes courts from awarding compensatory damages for physical injuries. In Australia, the professional associations cite the litigation cases when bargaining and when campaigning for system wide change to support their position (Australian Medical Association 1999). Self-regulation of working hours by regulatory councils/colleges or professional associations is minimal in these countries, most likely because of their dependence upon coregulation. Regulatory councils/colleges consider issues related to working hours in the context of disciplinary hearings into the practice of individual health professionals (e.g., Medical Board of Queensland v Doneman 2004). A limited example of voluntary self-regulation is the Australian Medical Association guidelines (as discussed above) (Australian Medical Association 1999). Coregulatory agreements through the process of collective bargaining are the primary instrument used to regulate working hours in these countries. The focus of these agreements is primarily, although not exclusively, to ensure occupational health and safety. These agreements are sector and profession specific, and involve a cooperative engagement between policy actors, generally professional associations, government, at the national or state/provincial level, or other government agents such as the District Health Boards in New Zealand. However, the negotiation processes have illustrated the tensions inherent in regulation where the parties may be in conflicted positions. In New Zealand, for example, physicians-in-training went on strike in 2006, claiming progress in reducing hours was too slow, probably due to government resource limitations (New Zealand Press Association 2006). In this case, those acting for government presumably wanted to improve occupational health and patient safety, as is its duty as an employer, yet these interests are in conflict with the overarching need to contain the national and regional budgets for health services. The effectiveness of these agreements remains somewhat of an open question. Generally, coregulatory instruments achieve simple effectiveness due to three factors: the prompting of a reevaluation of professional practice, a broad base of agreement between parties, and gradual implementation. In these countries, professional practice norms were reevaluated either as part of or prior to negotiation processes. The profession concluded that there were occupational health and safety issues for professionals, safety and quality issues for patients, and issues relating to adequacy of training associated with long working hours. The professions concluded that a reduction in working hours was acceptable or indeed desirable and that training could occur successfully within a reduced span of hours (the Royal Colleges in the United Kingdom concluded that training could be delivered successfully in fifty-six hours), a perspective that government agreed with. Because the professions accepted that change was required, there were few divisions between physicians and physicians-in-training, and in at least one country, New Zealand, the professional association representing senior physicians 128 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy was privy to collective bargaining negotiations. This is in contrast with the United States where no such reevaluation has occurred, and schisms remain within the medical profession about the need to reduce working hours. Lastly, coregulatory agreements have tended to result in gradual implementation of agreements enabling time for health providers to adjust practices by, for example, recruiting additional staff and changing rostering practices. However, coregulation is not without its problems, and so the picture is not entirely rosy. Certainly, there remains only sparse evidence about complex effectiveness. There is some limited evidence that occupational health improves and quality and safety of health services improves (Smakoff and Jacques 1991). Further, there have been problems with compliance. The experience in the United Kingdom, where, as discussed above, compliance with the “New Deal” agreement was approximately 80 percent indicates that compliance is difficult for health providers to achieve where there are shortages in human resources. There have also been problems with the agreements not achieving the desired reductions in working hours. New Zealand is not the only country where professional associations have expressed frustration. In Australia in 2002, the Australian Council of Trade Unions brought a test case, the Working Hours Case (2002), challenging mandatory overtime provisions on behalf of a number of collective agreements (including the Northern Territory Medical Officers) before the Australian Industrial Relations Commission.10 The Commission recognized that employees have the right to refuse to work overtime where it would result in the “working of unreasonable hours” (Working Hours Case: para. 287). It is not clear what impact this award right has had on working conditions for the employees in the collective agreements involved, or whether this has altered the health system’s employment practices at the macrolevel. After all, the court does not ban overtime, it merely creates a right for employees to refuse overtime—a right that some health providers, especially those in training, may not feel they can exercise, even with the backing of a union. It is also worthy of note that this award only addresses overtime—it does not address excessively long scheduled working hours. In these countries, the relative success of coregulation as an instrument to limit working hours was contingent on environments that facilitated such agreements. Coregulation through collective agreement has long formed part of the regulatory landscape in these countries. Unions have established relationships with their coregulatory partners and have some real influence in health systems built on accommodatory principles. Perhaps most importantly, the common characteristic of all these countries, and in the United Kingdom and Denmark when they used coregulatory schemes, is the greater involvement of government in the funding and provision of health services and, accordingly, governments’ influence in the negotiation of publicsector-focused collective agreements. MCDonald WORKING TO DEATH 129 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy C. EUROPEAN UNION AND NEW YORK STATE Regulation through litigation and command and control regulation are the primary instruments used to place limitations on working hours in the health systems in the European Union and in New York State. Recourse to regulation by litigation enables redress for those harmed,11 as changes to working practices as a result of litigation, if any, are invisible at a sectoral level, although they may impact specific health providers. In common with the other jurisdictions examined in this article, self-regulation plays a very limited role at the microlevel in relation to specific cases. The rationale for the introduction of command and control regulation in respect of this issue differed markedly between New York State and the European Union. New York introduced legislation to revise section 405 of the New York State Health Code in 1989 after the death of a patient in a New York hospital (New York State Health Code 2004).12 A grand jury examining the circumstances surrounding the death found fault with a system that routinely allowed physicians-in-training to work more than 100 hours per week, for continuous periods of thirty to forty hours. The grand jury found that overworked, sleep-deprived physicians-in-training, and a lack of supervision, were serious potential dangers for patients and that the methods for training physicians were “counterproductive to providing quality medical care” (New York Supreme Court 1986; Bell 2003). The New York legislation was a reactive response to a specific incident that highlighted patient safety issues. New York State is not alone in the United States in considering command and control regulation to address this issue. One U.S. territory, Puerto Rico13 also has legislation, a number of states are considering enacting similar legislation, including Massachusetts, Delaware, New Jersey, Pennsylvania, and California, and there have been, so far unsuccessful, federal legislative initiatives to address working hours issues in health care (Patient and Physician Safety and Protection Act of 2005; Safe Nursing and Patient Care Act of 2007). In contrast, the European Union focused on protecting occupational health and safety across all sectors; as a matter of policy, no employee working in any sector should be subject to the deleterious effects of working unduly long hours unless they voluntarily choose to. Limitations on working hours in the health sector were not the focus of regulation but a by-product of a general reform of working conditions. The European Commission passed the European Working Time Directive in 1993 (Council Directive 93/104/ EC). The UK parliament incorporated the Directive into law in England as the Working Time Regulations 1998. The Working Time Directive limited working time to a maximum of forty-eight hours per week by November 1996 for selected groups, including nurses and the medical profession, except physicians-in-training, but allowed for individual opt-outs. It subsequently became clear that on-call-hours count toward the working hour week when the European Court of Justice gave its rulings in Landeshauptstadt 130 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Kiel v Norbert Jaeger (2003) and Sindicato de Médicos de Asistencia Pública (Simap) v Conselleria de Sanidad y Consumo de la Generalidad Valenciana (2001). In 2000, the European Union amended the Working Time Directive (Council Directive 2000/34/EC) to include previously excluded groups. The United Kingdom passed the Working Time Amendment Regulations 2003, with application to physicians-in-training from 2004. The Working Hours Directive allowed for gradual implementation and advance notice for the regulated to enable planning for implementation. The welfare focus of the European Directive may explain why the European Union has the strictest limitations on working hours of all the countries examined in this article, with all health professionals being required to work a maximum of fifty-six hours per week reducing to forty-eight by 2009. These initiatives have received mixed reviews in terms of effectiveness. The most significant problem illustrated by the experiences in New York was a lack of compliance—there were no compliance mechanisms built into the original legislation. In 1998, the New York State Department of Health conducted a four-day unannounced investigation of twelve hospitals across New York State. All had violated the physician-in-training working hour limits. The department reported that: 37 percent of physicians-in-training worked more than eighty-five hours per week; 20 percent exceeded ninety- five hours per week; 60 percent of surgical residents exceeded ninety-five hours per week and 38 percent of physicians-in-training worked in excess of twenty-four hours per shift (Kwan and Levy 2004). Physicians-in-training also reported busy on-call time with limited rest (ibid.). In 2002, 66 percent of hospitals surveyed were not in compliance with the regulations. Fifty-six percent violated the twenty-four-hour shift requirements, 34 percent violated the eighty hours per week requirements, 23 percent did not provide twenty-four hours off, and 13 percent did not provide the required hours off between shifts (Lawrence 2003). Some felt that poor compliance with the New York law was due to the minor penalty, a maximum $2,000 fine, and a lack of enforcement by regulators. Figures from the United Kingdom also indicate that the Working Time Regulations made little impact upon the working hours of consultants and other physicians (MacDonald 2004). A recent study shows that one quarter of the working conditions of some positions conflict with the regulations, but there have been no prosecutions (ibid.). One concern is that the Health and Safety Executive does not appear to have enough inspectors or time to enforce the legislation (ibid.). When investigation revealed the deficits in the effectiveness of the New York law, government enacted the Health Care Reform Act of 2000 to strengthen the penalties to up to $50,000 for repeat violations. The Act included increased capacity in respect of enforcement and transparency measures. Subsequently the state hired an investigation agency to make yearly surprise visits to teaching hospitals, to issue an annual report, and to publish a yearly compliance report.14 In 2002, 60 percent of hospitals had one noncompliant program but most subsequently achieved compliance MCDonald WORKING TO DEATH 131 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy (Lawrence 2003). In New York, the reduction of the working hours of health professionals required the enactment of legislation, and, most importantly, it required that the law was enforced. The evidence in regard to complex effectiveness is also equivocal. Three studies examined the impact of the New York legislation. Researchers undertook a comparative study of the quality of care received by patients in October 1988 (when no work limitations were in place) and October 1989 (after work limitations were in place) (Laine et al. 1993). The study concluded that limited working hours were associated with delayed test ordering and increased complication rates, although these did not result in statistically significant differences in more serious outcomes, that is, in-hospital mortality, transfer to intensive care, discharge disposition, or length of stay. Another study, also in 1989, reported no difference in mortality, morbidity, or unexpected transfers after implementation of the section 405 amendments (Daiger, Welliver and Stapleton 1990). A third study noted reductions in mortality in 1991, but there were reductions in both teaching and nonteaching hospitals suggesting that the new section 405 law could not be responsible (Howard, Silber and Jobes 2004). As the legislation was still new when these studies were undertaken, some teething problems in transitioning to a position where physicians worked fewer hours was inevitable, so the results must be interpreted with some caution. In addition, many facilities ignored the legislation until 1998. In short, there appears no empirical evidence that patient safety improved or declined in New York State as a consequence of these reforms. Although the Working Hours Directive had at its base a concern for the health and safety of workers, its implementation in England was influenced by concerns about patient safety. A 2005 report from England suggests that a National Health Service response to the Working Hours Amendment Regulations “A Hospital at Night” (which was a model of shift patterns and staffing mix), delivered improvements to patient care (Mahon 2005). This was a qualitative assessment based on the perceptions of involved health professionals. The report’s authors noted that most health professionals said that they prioritized acutely ill patients presenting at night so that physicians that were more alert saw patients faster. Respondents also felt that there were fewer patient safety incidents, as there was more oversight of physicians-in-training by a senior nurse, and more health professionals attended to each patient. The report also noted that respondents felt that reducing hours had no detrimental effect on physicians’ training or on the achievement of national performance targets (ibid.). V. SCOPE OF REGULATORY LIMITATIONS ON WORKING HOURS One of the marked variances between the countries/supranational bodies examined during the course of this research was in the outcome of regulatory 132 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy processes to limit working hours. Every country or supranational body has established different parameters for working hours (Table 2). Comparing the extreme ends of the working hour spectrum, there is a forty-hour per week difference between the eighty-eight hours permitted in the United States (except for New York State) and the European Union’s goal of forty-eight hours (which will be in place by 2009). There is an eighteenhour difference in the maximum shift that physicians working in emergency departments in New York State may work compared with their colleagues elsewhere in the United States. Such significant differences are troubling given that the problem that these regulatory initiatives are addressing is the same. Overall, limits on working hours in North America (Canadian and U.S. schemas (the proposed federal, New York State, and ACGME’s national associational self-regulation)) are substantially higher than the limits in the other countries or the supranational body examined for the purposes of this research. So what possible explanations could account for these differences? A cursory examination would suggest that differences in regulatory imperatives are significant. The primary influence on the regulation of working hours in the United States was concern about patient safety; whereas working hours regulation elsewhere was driven primarily by concerns about occupational health and safety. However, as an explanation, this is inadequate, as the regulatory imperatives in Canada, where the working hours limitations are substantially similar to the United States, were primarily to improve occupational health. Additionally, as a coregulatory process established Canada’s working hours limitations, whereas the U.S. levels were established by Table 2. Regulatory Limitations on Working Hours of Physicians-in-Training Countries / Supranational Bodies Weekly hours Continuous duty time Days off Australia Varies from state to state Varies from state to state Varies from state to state Canada 70–80 hours 24 hours + handover 1 in 7 European Union (UK; Denmark) 58 hours (reducing to 48 hours by 2009) 13 hours 1 in 7 or 2 in 14 New York State 80 hours (averaged over 4 weeks) 12 hours (Emergency) 24 hours all other specialties 1 in 7 New Zealand 56–58 hours 16 hours 1 in 7 United States (except NY) 80 hours (averaged over 4 weeks with 10 percent increase providing sound educational rationale) 30 hours (24 duty + 6 hours for continuity of care and educational activities) 1 in 7 (averaged over four weeks) MCDonald WORKING TO DEATH 133 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy legislation in New York State and associational self-regulation in the rest of the United States, instrument choice does not appear to have dictated similarities in regulatory outcomes. Instead, two factors may determine outcomes: a desire for consistency between neighboring jurisdictions; and perhaps that the culture of the medical profession in North America differs substantively from that in Europe and Australasia. While cultural issues are notoriously difficult to identify, the rhetoric surrounding working hours seems to point to three substantive points of difference in the cultures of medicine in these countries. The first point of difference seems to be in the interpretation of the nature of the ethical obligations owed to patients. One perspective considers that a professional’s obligations to patients can best be served if the professional is mindful of his or her physical limitations. The other perspective, common in the United States, appears to be that the patient’s interests are best served through continuity in care, no matter what physical or psychological burden this places on the health provider. The second point of difference relates to the nature of the process of training physicians. Quality rather than quantity appears the guiding principle in Europe, whereas quantity equals quality could be characterized as the position held by many in the United States. The third point of difference appears to be differing contexts of professional practice. Government and the professions work closely to ensure the functioning of the health system in most of the countries examined in this research, with the exception of the United States. In the acute sector in Australia, New Zealand, Denmark, and the United Kingdom, many physicians are salaried employees of government or government agencies. In Canada, governments fund health services, but physicians are independent contractors earning fee for service payments. In the United States, the provision of health services by physicians is even more commercial in orientation. The medical profession’s sense of autonomy in its professional practice in non-North American countries may be tempered by an appreciation of the nature of relationships within the health system and with government and government actors. In contrast, the relative autonomy of the medical professions in North America may be significant in explaining why the professions are reluctant to place greater limitations on working hours. VI. CONCLUDING COMMENTS The consensus among all the countries and the supranational body examined in this review is that they must place some limits on the hours worked by health professionals, especially physicians-in-training. They also concur, some more recently than others, that regulation by litigation as a single instrument is insufficient to serve the public interest in assuring the health and safety of health professionals, patients, and the public more generally. Thus, they must supplement regulation by litigation with other instruments 134 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy to achieve desired policy outcomes. It is at this point that the consensus ends, as different jurisdictions use different supplementary instruments. What does seem clear from the experiences of the jurisdictions examined in this article is that the sociopolitical cultures of the countries are a determinant of instrument choice. The majority of the countries examined in this report have systems that government largely funds and/or administers. Implementing limits on working hours by coregulation through collective bargaining appears to be the instrument of choice for this cohort. In these countries, health professionals are unionized, and negotiations in respect of employment conditions are part of the fabric of these health systems. Denmark and the United Kingdom only departed from this model due to their supranational obligations. The European Union requires consistency between member states and is committed to reforming work environments to ensure health and safety. In contrast, government’s role as a funder or administrator is less pervasive in the United States, and collective negotiations are not part of the fabric of its health systems with their more individualized and profit driven focus. Thus, instrument choice is limited to encouraging self-regulation, generally through threatening professional autonomy or by imposing command and control regulation. Further, the experiences of the countries under review indicate that perceptions of professional practice norms are a determinant of instrument effectiveness, at least in simple terms. In all countries but the United States, there is consensus within the profession that lower hours of work are appropriate or indeed required to address health and safety concerns and that the training of health professions is not adversely affected, and may be improved, by imposing such limits. In contrast, the debate rages in the United States as to whether limits on working hours are required. Within the professions in the United States, there is dissension about conflicting ethical obligations, issues in relation to training, and in respect of what is good practice. The degree of resistance experienced in the United States to self-regulation, and indeed to command and control regulation, appears to have its roots in dissension rather than pure resource constraints. The U.S. experience indicates that a regulatory instrument will be undermined where the aims of the regulators do not have the broad support of the profession and where individuals are confronted with internal and external pressures and expectations to act contrary to those aims. These factors may be as intangible but powerful as professional culture or as tangible as human resource shortages that mean that facilities are simply not able to recruit sufficient staff to comply with the limitations. Regulation may achieve complex effectiveness in terms of the safety and well-being of workers; however, its record in respect of improving the quality and safety of health care is equivocal. This is because of the interrelated processes that comprise patient care so that when one element changes (i.e., physician working hours are reduced) other processes must also change, otherwise the safety and quality of the care will be reduced. Rigorous internal MCDonald WORKING TO DEATH 135 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy processes and innovation must accompany regulations to reduce working hours to ensure that patient safety and quality care are, at the very least, not compromised, if not enhanced. As a final comment, the experiences of these countries in the regulation of working hours indicate the challenges faced by regulators but also by the regulated. At the macrolevel, regulators balance the public interest in protecting the safety of patients and the health and safety of health professionals against the very real and equally compelling issues of resource management in the health system. Perhaps the most notable resource issue is the international scarcity of health professionals, a problem projected to grow worse as the century, even the decade, progresses. At the microlevel, health professionals’ ethical and moral obligations to do no harm and their interest in ensuring that workplace conditions promote health and safety are counterbalanced by issues about the nature of professional service to patients and economic interests. A fine balance indeed. fiona mcdonald is an Associate Lecturer in the School of Law at Queensland University of Technology, Australia. She previously worked as a Research Associate at the Health Law Institute at Dalhousie University, Canada, and her research interests encompass regulation and patient safety. NOTES 1. This research is part of a larger project that examined the governance of patient safety in six countries (Downie et al. 2006). The methods employed were a search of the literature, grey literature and legislation, regulation, contracts, and collective agreements. Informational interviews with key agencies from the countries examined were also conducted to confirm instrument usage and impact. Research ethics approval was sought but the Research Ethics Office at Dalhousie University deemed that ethics review was not required. 2. It is important to acknowledge that a reduction in the hours worked by health professionals is not the sole remedy for patient safety-related issues in health systems. It is rather one solution among many proposed and initiated by policy actors. Examples of other strategies to improve patient safety include: in the United States a significant investment in research on safe clinical delivery and in the United Kingdom the introduction of a clinical governance framework (see Downie et al. 2006). 3. New Zealand and Denmark have no-fault insurance schemes that discourage negligence claims. 4. The Supreme Court of Illinois refused leave to appeal in 2006 (Brewster v RushPresbyterian St. Luke’s Medical Center). 5. ACGME is a private professional organization responsible for the accreditation of medical training programs—it is a pure self-regulatory body in that government does not officially sanction its operations, functions, and powers through the passage of empowering legislation. 6. The U.S. government spends about $8 billion per year to train residents. 7. A federal bill placing restrictions on the hours nurses may work in federally funded institutions was also introduced at the same time as the bill relating to 136 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy physicians-in-training. The latest incarnations of these bills were the Patient and Physician Safety and Protection Act of 2005, which lapsed at the committee stage and the Safe Nursing and Patient Care Act of 2007, which has recently been reintroduced after a previous bill lapsed. 8. One in five failed to comply with duty hour requirements in 1999 with compliance improving slightly in 2000. 9. The Yale surgical program lost accreditation in 2001. It regained (provisional) accreditation in 2002. 10. The Australian Industrial Relations Commission is a national tribunal dealing with employment issues. 11. Denmark has a no-fault system for harms caused by treatment received in the health system. 12. Libby Zion was an eighteen-year-old woman who died in a New York hospital in 1984. Her father claimed she had received inadequate care from overworked and undersupervised junior medical staff and that this caused her death. 13. An Act to Regulate the Work Shifts of Medical Interns and Residents in Puerto Rico (2003) P.R. Laws 47. An English translation is available online at American Medical Student Association, http://www.amsa.org/hp/rwh_pr.doc (accessed 10 July 2007). 14. New York State closed down Mt Sinai’s living donor adult liver transplant program for one year after a patient died citing breaches of the 405 regulations (especially in regard to supervision and staffing levels). Mt Sinai lost $70 million (U.S.) as a consequence. REFERENCES Abraham, Kenneth (2002) “The Insurance Effects of Regulation by Litigation.” In Regulation through Litigation, edited by W. K. Viscusi. Washington, DC: AEIBrookings Joint Center for Regulatory Studies. Accreditation Council for Graduate Medical Education (2003) Resident Duty Hours Standards. Available at: http://www.acgme.org/acWebsite/dutyHours/dh_dutyHoursCommonPR.pdf. Accreditation Council for Graduate Medical Education (2006) “Most Residents Followed Weekly Duty Hour Limits in 2005–06,” News Release/Communiqué 3 November. Aiken, L., and D. Sloane (2002) “Hospital Organisation and Culture.” In Hospitals in a Changing Europe, edited by M. McKee & J. Healy. Buckingham, Philadelphia: Open Univ. Press. Altman, L. K., and D. Grady (2002) “Hospital Accreditor will Strictly Limit Hours of Residents,” New York Times 13 June: A-1. American Medical Student Association (2005) “Study Proves Work Hour Guidelines for Interns are Unsafe,” News Release/Communiqué 17 January. Annas, G. (2006) “The Patient’s Right to Safety—Improving the Quality of Care through Litigation against Hospitals,” New England Journal of Medicine 354: 2063–66. Anonymous (2002) “Sleep Deprived Doctors,” New York Times 14 June: A-36. Australian Medical Association (1999) National Code of Practice—Hours of Work, Shiftwork and Rostering for Hospital Physicians. Canberra, ACT: Australian Medical Association. Australian Medical Association (1998) Review of Overseas Experience in Regulating Hours of Work of Physicians in Training. Kingston, ACT: Australian Medical Association. Bailit, Jennifer, and May Blanchard (2004) “The Effect of House Staff Working Hours on the Quality of Obstetric and Gynecologic Care,” Obstetrics and Gynecology 103: 613–16. MCDonald WORKING TO DEATH 137 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Barden, C. B., M. C. Specht, M. D. McCarter, J. Daly, and T. Fahey III (2002) “Effects of Limited Work Hours on Surgical Training,” Journal of the American College of Medicine 195: 531–38. Barger, L., B. Cade, N. Ayas, J. Cronin, B. Rosner, F. Speizer, C. Czeisler, Harvard Work Hours, Health, and Safety Group (2005) “Extended Work Shifts and the Risks of Motor Vehicle Crashes Among Interns,” New England Journal of Medicine 352: 125–34. Bell, Bertrand (2003) “Reconsideration of the New York State Laws Rationalizing the Supervision and the Working Conditions of Residents,” Einstein Journal of Biology and Medicine 20: 36–40. Boodman, S. (2001) “Limited Experience: High Expectations,” Washington Post 27 March: H-01. Braithwaite, John (2002) “Rewards and Regulation,” Journal of Law and Society 29: 12–26. Braithwaite, J., J. Healy, and K. Dwan (2005) The Governance of Health Safety and Quality. Canberra: Commonwealth of Australia. Brennan, Troyen, and Michael Zinner (2003) “Residents’ Work Hours: A Wake Up Call?,” International Journal of Quality in Health Care 15: 107–08. British Medical Association (2006) Briefing Note: Hospital Doctors—Junior Doctors Hours. London: British Medical Association. British Medical Association (1999) Report on the Health and Safety Implications Associated with Physicians Working Patterns. London: British Medical Association. British Medical Association Health Policy and Economic Research Unit (1998) Work Related Stress Among Junior Physicians. London: British Medical Association. Brunworth, J. D., and R. Sindawi (2006) “Impact of Duty Hour Restrictions on Otolaryngology Training: Divergent Resident and Faculty Perspectives,” Laryngoscope 116: 1127–30. Buchan, James (2004) A Certain Ratio? Minimum Staffing Ratios in Nursing: A Report for the Royal College of Nursing. London: Royal College of Nursing. Butterfield, S. (2007) “Resident’s Union Celebrates its Golden Anniversary,” ACP Observer July/August. Available at: http://www.acponline.org/journals/news/jul-aug07/union.htm. Carpenter, R. O. M. Austin, J. Tarpley, M. Griffin, and K. Lomis (2006) “Workhour Restrictions as an Ethical Dilemma for Residents,” American Journal of Surgery 191: 527–32. Cass, Hilary, Isabel Smith, Cheryl Unthank, Colin Starling, Jane E. Collins (2003) “Improving Compliance with Requirements on Junior Doctors Hours,” British Medical Journal 327: 270–73. Child, Stephan, and Andrew Old (2004) “Resident Medical Officer Working Conditions in New Zealand: Results of a Recent Survey,” New Zealand Medical Association Journal 117: 1–11. Cohen-Gadol, A., D. G. Piepgras, S. Krishnamurthy, and R. D. Fessler. (2005) “Resident Duty Hours Reform: Results of a National Survey of the Program Directors and Residents in Neurosurgery Training Programs,” Neurosurgery 56: 398–403. Croasdale, Myrle (2004) “Resident Work-hour Limits still a Struggle One Year into Restrictions: General Compliance Appears to be the Norm, But Residents See Flaws in the System,” AMN News 19 July. Available at: http://www.amaassn.org/amednews/2004/07/19/prl10719.htm. Daiger, G. E., R. C. Welliver, and F. B. Stapleton (1990) “New York Regulation of Residents’ Working Conditions; 1 Year’s Experience,” American Journal of Diseases in Children 144: 799–802. Danzon, P. (1985) Medical Malpractice: Theory Evidence and Public Policy. Cambridge, MA: Harvard Univ. Press. Day, P., and R. Klein (1987) “The Regulation of Nursing Homes: A Comparative Perspective” The Milbank Quarterly 65 (3): 303–47. 138 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Department of Health, National Assembly for Wales, the NHS Confederation and the British Medical Association (2002) Guidance on Working Patterns for Junior Physicians. London: Department of Health. Downie, J., W. Lahey, D. Ford, E. Gibson, M. Thomson, T. Ward, F. McDonald, and A. Shea (2006) Patient Safety Law: From Silos to Systems: A Report to Health Canada. Ottawa: Health Canada. Eastman, Nigel, and Jill Peay eds. (1999) Law without Enforcement: Integrating Mental Health and Justice. Oxford: Hart. Fischer, Josef (2004) “Continuity of Care: A Casualty of the 80-hour Work Week,” Academic Medicine 79 (4): 381–83. Fletcher, K. E., S. Q. Davis, W. Underwood, R. S. Mangrulkar, L. F. McMahon Jr., and S. Saint (2004) “Systemic Review: Effects of Resident Work Hours on Patient Safety,” Annals of Internal Medicine 141: 851–57. Gotbaum, Rachel (2005) “Safety of Medical Residents’ Long Hours Questioned,” National Public Radio 28 February. Available at: http://www.npr.org/templates/story/story.php?storyId=4512366. Grabosky, P. (1995) “Using Non-Governmental Resources to Foster Regulatory Compliance,” Governance: An International Journal of Policy and Administration 8: 527–50. Green, M. (1995) “What (if Anything) Is Wrong with Residency Overwork?,” Annals of Internal Medicine 123: 512–17. Gunningham, N., and P. Grabosky (1998) Smart Regulation: Designing Environmental Policy. Oxford: Clarendon/Oxford Univ. Press. Gunningham, N., and R. Johnstone (1999) Regulating Workplace Safety: Systems and Sanctions. Oxford: Oxford Univ. Press. Hexom, Braden (2004) “Prepared Statement for the New Jersey Assembly in Support of Assembly Bill 439 An Act Concerning Safe Resident Staffing Standards Regulated Professions and Independent Authorities Committee,” American Medical Student Association 3 May. Available at http://www.amsa.org/rwh/NJ.cfm (accessed 10 July 2007). Howard, D. L., J. H. Silber, and D. R. Jobes (2004) “Do Regulations Limiting Residents’ Work Hours Affect Patient Mortality?,” Journal of General Internal Medicine 19: 1–7. Hutter, M. M., K. C. Kellogg, C. M. Ferguson, W. M. Abbott, and A. L. Warshaw (2006) “The Impact of the 80-Hour Resident Workweek on Surgical Residents and Attending Surgeons,” Annals of Surgery 243: 864–75. Jacobson, P., and Soheil Soliman (2002) “Litigation as Public Health Policy: Theory or Reality?,” Journal of Law, Medicine & Ethics 30: 224–38. Kagan, Robert A., and John T. Scholz (1984) “The ‘Criminology of the Corporation’ and Regulatory Enforcement Strategies.” In Enforcing Regulation, edited by K. Hawkins & J. M. Thomas. Boston: Kluwer-Nijhoff. Kwan, Rita, and Robert Levy (2004) A Primer on Resident Work Hours, 5th ed. Reston, VA: American Medical Student Association. Laine, C., L. Goldman, R. Soukup, and J. G. Hayes (1993) “The Impact of a Regulation Restricting Medical House Staff Working Hours on the Quality of Care,” JAMA 269: 374–78. Lawrence, Hal (2003) “The Impact of Residents’ Work Hour Restrictions,” Current Women’s’ Health Reports 3: 487–91. Institute of Medicine (2000) To Err Is Human: Building a Safer Health System. Washington DC: National Academies Press. Institute of Medicine (2004) Keeping Patients Safe: Transforming the Work Environment of Nurses. Washington DC: National Academies Press. MacDonald, Rhona (2004) “How Protective is the Working Time Directive?” British Medical Journal 329: 301–02. Mahon, Claire Harris, Jane Tyrer, Sarah Carr, Karin Lowson, Louise Carr, Stephen Chaplin, and Dianne Wright (2005) The Implementation and Impact of Hospital at Night Pilot Projects: An Evaluation Report. London: Department of Health. MCDonald WORKING TO DEATH 139 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Mello, M., and T. Brennan (2002) “Deterrence of Medical Errors: Theory and Evidence for Malpractice Reform,” Univ. of Texas Law Review 80: 1595–1637. Mello, M., C. Kelly, and T. Brennan (2005) “Fostering Rational Regulation of Patient Safety,” Journal of Health Politics, Policy and Law 30 (3): 375–426. New Zealand Press Association (2006) “Doctors Vote to Strike Over Hours” New Zealand Herald: 31 May. New York Supreme Court, County of NY: Part 50 (1986) “Report of the Fourth Grand Jury for the April/May Term of 1986 Concerning the Care and Treatment of a Patient and the Supervision of Interns and Junior Residents at a Hospital in New York County.” Nocera, Anthony, and Diana Strange Khursandi (1998) “Physicians’ Working Hours: Can the Medical Profession Afford to Let the Courts Decide What is Reasonable?” Medical Journal of Australia 168: 616–18. Ogus, A. (1995) “Rethinking Self-Regulation,” Oxford Journal of Legal Studies 15 (1): 97–108. Okie, Susan (2007) “An Elusive Balance—Residents’ Work Hours and the Continuity of Care,” New England Journal of Medicine 356: 2665–67. Parshurum, C., S. Dhanani, J. Kirsh, and P. Cox (2004) “Fellowship Training, Workload, Fatigue and Physical Stress: A Prospective Observational Study,” Canadian Medical Association Journal 170: 965–70. Patton, David, Dawn R. Landers, and Indu Agarwhal (2001) “Legal Considerations of Sleep Deprivation Among Resident Physicians,” Journal of Health Law 34 (4): 377–417. Pickersgill, T. (2001) “The European Working Time Directive for Doctors in Training,” British Medical Journal 323: 1266. Pilcher, J., and A. Huffcutt (1996) “Sleep Deprivation on Performance: A MetaAnalysis,” Sleep 19: 318–26. Rhodes, R. (1997) Understanding Governance: Policy Networks, Governance, Reflexivity and Accountability. Buckingham, UK: Open Univ. Press. Romanchuk, K. (2004) “The Effect of Limiting Residents’ Work Hours on Their Surgical Training: A Canadian Perspective,” Academic Medicine 79: 384–85. Sakorafas, G., and G. G. Tsiotos (2004) “New Legislative Regulations, Problems, and Future Perspectives, with a Particular Emphasis on Surgical Education,” Journal of Postgraduate Education 50: 274–77. Salamon, Lester ed. (2002) The Tools of Government. New York: Oxford Univ. Press. Scott, W. D., and Deloitte Ltd. (1987) Final Report for the Department of Health on Current Medical Officer Rostering Practices. Wellington, New Zealand. Cited by Judith Smakoff and C. M. Jacques (1991) “A Review of Studies Concerning Effects of Sleep Deprivation and Fatigue on Residents Performance,” Academic Medicine 66: 687–93. Sharpe, Virginia A. (2004) “Accountability and Justice in Patient Safety Reform.” In Accountability: Patient Safety and Policy Reform, edited by V. A. Sharpe. Washington DC: Georgetown Univ. Press. Smakoff, Judith, and C. M. Jacques (1991) “A Review of Studies Concerning Effects of Sleep Deprivation and Fatigue on Residents Performance,” Academic Medicine 66: 687–93. Stamp, T., P. Termuhlen, S. Miller, D. Nolan, Hutzel, J. Gilchrist, and R. Johnson (2005) “Before and After Resident Work Hour Limitations: An Objective Assessment of the Well-being of Surgical Residents,” Current Surgery 62 (1): 117–21. Steinbrook, R. (2002) “The Debate over Residents’ Work Hours,” NEJM 347: 1296–1302. Tuohy, Carolyn (1999) Accidental Logics: The Dynamics of Change in the Health Care Arena in the United States, Britain and Canada. New York: Oxford Univ. Press. Walshe, K. (2003) Regulating Healthcare a Prescription for Improvement. Maidenhead, UK: Open Univ. Press. 140 LAW & POLICY January 2008 © 2008 The Authors Journal compilation © 2008 Baldy Center for Law and Social Policy Viscusi, W. Kip (2002) “Overview.” In Regulation through Litigation, edited by W. K. Viscusi. Washington DC: AEI-Brookings Joint Center for Regulatory Studies. United Kingdom, H. C. Hansard, 6th ser., vol. 412, col. 639–644 (3 November 2003). CASES CITED Australia Brotherston v Royal Perth Hospital (1995) 15 SR (WA) 42a. Medical Board of Queensland v Doneman [2004] QHPT 010. Working Hours Case (July, 2002) PRO72002 AIRC. European Court of Justice Landeshauptstadt Kiel v Norbert Jaeger [2003] ECJ Case C-151/02. Sindicato de Médicos de Asistencia Pública (Simap) v Conselleria de Sanidad y Consumo de la Generalidad Valenciana [2001] ECJ Case C-303/98, ICR 1116. United Kingdom Johnstone v Bloomsbury Health Authority [1991] 2 All ER 801. Wilsher v Essex Health Authority [1986] 3 All ER 801. United States Boston Medical Center Corporation v House Officers’ Association/Committee of Interns and Residents, 330 NLRB 152 (1999). Brewster v Rush-Presbyterian-St. Luke’s Medical Center, 361 Ill App 3d 32; 836 NE 2d 638 (Ill App 2005). Brewster v Rush Presbyterian-St Luke’s Medical Center, 217 Ill 2d 598; 844 NE 2d 964 (Ill SC 2006). STATUTES CITED European Union Council Directive 2000/34/EC of the European Parliament and of the Council of 22 June 2000 amending Council Directive 93/104/EC concerning certain aspects of the organization of working time to cover sectors and activities excluded from that Directive O.J. L.195/41. Council Directive 93/104/EC of 23 November 1993 concerning certain aspects of the organization of working time, O.J. L.307/18. United Kingdom Working Time Regulations 1998, SI 1998/1833. The Working Time (Amendment) Regulations 2003, SI 2003/1684. United States An Act to Regulate the Work Shifts of Medical Interns and Residents in Puerto Rico, P.R. Laws 47 (2003). Health Care Reform Act of 2000, N.Y. New York State Health Code, N.Y. Comp. Codes & Regs tit. 10, § 405.4 (2004). Patient and Physician Safety and Protection Act of 2005, Bill HR 1228, 109th Cong., 2005. Safe Nursing and Patient Care Act of 2007, Bill HR 2122, 110th Cong., 2007. 

AccountingQueen

(3)

Status NEW Posted 07 Aug 2017 08:08 AM My Price 10.00

Hel-----------lo -----------Sir-----------/Ma-----------dam----------- T-----------han-----------k y-----------ou -----------for----------- us-----------ing----------- ou-----------r w-----------ebs-----------ite----------- an-----------d a-----------cqu-----------isi-----------tio-----------n o-----------f m-----------y p-----------ost-----------ed -----------sol-----------uti-----------on.----------- Pl-----------eas-----------e p-----------ing----------- me----------- on----------- ch-----------at -----------I a-----------m o-----------nli-----------ne -----------or -----------inb-----------ox -----------me -----------a m-----------ess-----------age----------- I -----------wil-----------l

Not Rated(0)

• Buy Answer