The policy push for greater transparency about adverse events in health care is at odds with the traditional risk-management view that such revelation increases the exposure of clinicians and health care institutions to liability.1 Open disclosure (OD), which relies heavily on the active involvement of health professionals, brings this tension to a head. Surveys of doctors in the United States have consistently identified fear of increased liability as one of the main reasons they are reluctant to communicate with patients about adverse events.2,3 Findings from a recent survey in Australia suggest that the same may be true here.4
Are these fears well founded? Unfortunately, no strong empirical evidence exists to determine whether openly communicating about unanticipated outcomes of care increases clinicians’ exposure to medicolegal activity, reduces it, or leaves it unaffected.2 There are theoretical reasons to expect that OD may stimulate litigation.5 On the other hand, there are no signs of spikes in medical negligence litigation, health care complaints or medical board actions in Australia over the past 5 years as interest in OD has grown. Moreover, several institutions in the US that have adopted progressive disclosure policies have reported favourable experiences.6-8
Evaluating the impact of OD is complex: these communications are difficult to track, and the long tail of litigation means that rigorous population-level evidence of the effects on patients’ claim and complaint-lodging behaviour is unlikely to emerge any time soon.9 In the meantime, given the apparent concerns of health professionals in this area,4,10 it is useful to examine what legal protections exist in Australia to guard against the medicolegal use of information conveyed in OD by patients, courts and regulators.
We reviewed Australian law in this area, focusing on apology laws and qualified privilege, the two main bodies of relevant law.
The national Open Disclosure Standard (“National Standard”) includes an “expression of regret” as an appropriate element of a disclosure, defining it as “an expression of sorrow for the harm experienced by the patient”.11 All states and territories have “apology laws” — statutory provisions that protect statements of apology or regret made after “incidents” from subsequent use in various legal contexts. These laws were not enacted with OD in mind; they apply to a much broader range of activities and pre-date the OD movement.
Across jurisdictions, apology laws share some common features, but variation occurs in four key areas (Box 1), affecting the scope and strength of these laws.
One notable point of variation is the definition of the apology or expression of regret to which the protections apply. Five jurisdictions (Queensland, Victoria, Tasmania, Western Australia, the Northern Territory) explicitly exclude statements containing acknowledgements of fault or liability, and a sixth jurisdiction (South Australia) does so implicitly. In other words, the protections are lost if the apology includes a mea culpa statement. Whether this definitional feature seriously undercuts the force of apology laws is questionable. The National Standard11 explicitly warns against making statements that admit liability for patient harm, and we see no sound rationale for doing so.
Another point of variation concerns the type of proceedings to which the shield applies. States and territories may protect apologies from use in civil proceedings of any kind (Tas, WA), exclude certain types of civil liability claims (New South Wales, the Australian Capital Territory) or limit the protection to proceedings in tort (SA) or personal injury damages claims (Qld, Vic, NT). Only the Victorian legislation explicitly covers medical board hearings.
An expression of regret, as the National Standard makes clear, is just one aspect of a comprehensive disclosure. Therefore, even when apology law protections apply and cover this aspect of the OD, they do not necessarily touch other aspects. For example, statements explaining the event or its causes are likely to be construed as separate and distinct from the apology, and thus remain discoverable and admissible in legal proceedings.
In sum, apology laws guard against the use of certain parts of OD conversations against health professionals in legal proceedings. As a shield, however, they are neither large nor thick. Their variability across jurisdictions and inherently selective nature (in addressing only one element of OD) limit their reach. For more wide-ranging legal protections, it is necessary to look elsewhere.
Qualified privilege is a legal doctrine that protects certain documents and communications from demands to disclose them in legal proceedings. The policy rationale is to encourage candidness and the free flow of information in circumstances in which that is regarded as socially beneficial. Governments in Australia have long taken the view that information produced as part of activities aimed at improving quality of care in health care institutions may warrant qualified privilege.
Statutes in four states (Qld, NSW, Vic, Tas) and both territories (ACT, NT) anchor qualified privilege in the “quality assurance committee” (QAC). The WA statute refers to a “quality improvement committee”,12 and the wording of a new law in SA focuses on the group undertaking an “authorised quality improvement activity”.13 NT law allows for qualified privilege only in relation to committees evaluating the quality of mental health services.14 The QAC, or its equivalent, is defined as a body engaged in quality assurance work and formally declared by the relevant minister as enjoying qualified privilege.
An important threshold question is how, if at all, qualified privilege relates to information conveyed in OD. Quality assurance work and OD are essentially different activities, but they may intersect. The main point of intersection relates to products of investigations, such as root cause analyses (RCAs),15 into the causes of unexpected outcomes of care.
In many hospitals, QACs undertake such investigations. The same adverse event on which a QAC is focused may be (or become) the subject of an OD. Optimal OD practice involves sharing with affected patients information about what happened and what is being done to prevent recurrences.11 Hence, elements of the OD communication may resemble or approximate information that a QAC later gathers and analyses. Flipping the timeline, when QAC findings become available, good OD practice may dictate that some or all of the findings be shared with affected patients and their families.
To what extent, then, does the privileging of information in the QAC setting radiate out to affect use of the information in the OD setting? On closer analysis, there are two distinct questions here: (1) Does qualified privilege law introduce prohibitions or barriers to the release of the information to patients through the OD process, based on the fact that such information may also fall within the ambit of a QAC’s work?; and (2) Does the protection qualified privilege laws provide to the QAC information extend to cover the same or similar information when it crops up in OD?
The statutory privileges that apply to quality assurance activities (in all states) and RCAs (in Qld, NSW, SA) do not pose a substantive legal barrier to full and candid OD. Perceptions that they do are not rooted in a firm understanding of the law. The main reason is that conventional OD practice, as outlined in the National Standard, cannot be construed as a quality assurance activity under any of the state or territory statutes. The fact that the incident that triggers the OD is or may become the subject of quality assurance work should not bar health professionals from sharing with patients information they have about what happened and why.
However, there is a more targeted version of the barrier question that is harder to answer definitively, namely: can the actual information produced by a QAC or RCA team be lawfully fed back to patients through an OD process? Looking across the state and territory statutes, there is no neat answer. It depends. Every jurisdiction imposes restrictions on the ability of persons outside QACs to demand information, and also on the ability of committee members to share it. In general, direct transfer of information from QACs to patients will be problematic. However, every jurisdiction also has lawful pathways that would enable the release of such information to patients to whose care the information relates. The more pertinent question, from a barriers perspective, is likely to be non-legal: how motivated are the involved clinicians, QACs, hospitals and health services to find and use those lawful pathways to facilitate the release of protected information to patients?
The primary purpose of qualified privilege laws is to protect against demands from outsiders — chiefly courts and other legal entities (eg, coroners, complaints commissions and insurers) — to share information generated in the course of quality assurance activities. The state and territory statutes articulate these protections in slightly different ways (Box 2).
In general, the protections are quite robust. From an OD perspective, however, the pivotal issue is the degree to which these protections extend beyond their home ground of quality assurance activities to cover information conveyed in OD. Our analysis suggests the answer is “minimally”, if at all. It is difficult to connect the protections conferred by qualified privilege to OD activities. This is true even for the recently enacted quality assurance provisions in Qld16 and SA,17 which add to the standard set of quality assurance provisions new protections relating specifically to information generated by RCAs.
Three main factors disrupt the applicability of qualified privilege protections to information in the OD setting. The first and most compelling factor relates to the ambit of the protections. We reiterate a key point made earlier and evidenced in Box 2 structurally, qualified privilege is fixed to QACs and their membership. There is some limited extension to a wider set of quality assurance activities — most notably in the federal law.18 But OD cannot easily be shoehorned into any of these “covered” categories. Moreover, OD activities were almost certainly outside the considerations of ministers at the time they declared particular entities to be eligible for qualified privilege.
Second, once a person obtains via a lawful pathway information that originated in a protected setting, such as a QAC, the protections that applied at its point of origin tend to fall away. WA is the only jurisdiction with protections that adhere strongly to the information once it is transferred lawfully to third parties.19
Third, the foregoing analysis, by focusing on the structural features of qualified privilege law, risks missing several practical reasons why this body of law may have little or no traction in the real-world context of OD practice. For example, many adverse events do not become the focus of work by a QAC or RCA team,11 rendering the qualified privilege laws essentially irrelevant. In addition, the principal focus of QAC and RCA activities is on systemic causes of adverse events. Elements of an OD communication that lack a connection to this focus will be untouched by qualified privilege protections. Finally, OD processes are usually initiated in the immediate aftermath of an event, well before it has time to become the formal subject of a QAC’s work. Thus, real-world temporal realities may obviate the applicability of privilege laws to much of what will be conveyed in OD.
In sum, the connection between state qualified privilege laws and OD processes is relatively weak, a weakness that cuts in two directions. On the one hand, it means that qualified privilege laws do not pose major barriers for standard OD practice. Indeed, in most jurisdictions, provided appropriate pathways for the release of information are followed, qualified privilege probably does not even bar feeding into the OD process information that comes out of quality assurance and RCA activities. On the other hand, the corollary of this weak connection is that existing qualified privilege laws do not provide robust protections over the content of OD conversations.
There is an important ground-shift occurring in health care systems globally toward greater openness with patients about adverse events.1 Whether OD will increase medicolegal risks for health professionals and hospitals remains to be seen, but many health professionals appear to think it will,4 a perception that likely inhibits the uptake of OD. Unfortunately, current Australian law does little to counteract that perception.
In our view, the situation presents a strong case for law reform. Ideally, such reform would:
provide strong, clear and reliable protections against use of the contents of OD conversations in subsequent legal proceedings;
clarify that qualified privilege law does not obstruct health professionals’ ability to conduct OD;
be broadly consistent across jurisdictions, in line with wider trends in clinical practice and national workforce regulation; and
be accompanied by an effective outreach effort to educate health professionals about what the new laws say.
Law reform along these lines would serve several important ends. First and foremost, it would be a clarion signal from policymakers that OD, like quality assurance, is a socially valuable activity. Second, new legislation could provide guidance on the important issue of whether and how information generated in QACs and RCAs may be lawfully fed into OD communications. Third, in a perfect world, medical ethics should be sufficient to drive health professionals’ commitment to OD, but the reality is that comfort on the medicolegal front is likely to prove a useful carrot: it should assuage the concerns of health care professionals and their indemnity insurers about the risks of undertaking OD after adverse events in health care.
1 Four key dimensions of state and territory apology laws, by jurisdiction
2 Form of restrictions by which qualified privilege laws protect information from use in legal proceedings, by jurisdiction
- David M Studdert1
- Mark W Richardson2
- 1 Melbourne School of Population Health and Melbourne Law School, University of Melbourne, Melbourne, VIC.
- 2 Aickin Chambers, Melbourne, VIC.
Our study was funded by the Australian Commission on Safety and Quality in Health Care. David Studdert was also supported by an Australian Research Council (ARC) Federation Fellowship. We thank Tamara Vu for excellent research assistance.
David Studdert is the chief investigator of an ARC linkage grant in which he is partnering with the Health Services Commissioner of Victoria and Avant Mutual to study complaints and claims regarding informed consent. Most of the project costs are met by the ARC, but Avant Mutual also contributes to the project budget. Mark Richardson is a member of the Victorian Bar and provides legal representation and advice to plaintiffs and defendants in a wide range of health-related matters.
- 1. Studdert DM, Brennan TA. No-fault compensation for medical injuries: the prospect for error prevention. JAMA 2001; 286: 217-223.
- 2. Gallagher T, Studdert DM, Levinson W. Disclosing harmful medical errors to patients: recent developments and future directions. N Engl J Med 2007; 356: 2713-2719.
- 3. Lamb RM, Studdert DM, Bohmer RMJ, et al. Hospital disclosure practices: results of a national survey. Health Aff (Millwood) 2003; 22: 73-83.
- 4. Studdert DM, Piper D, Iedema R. Legal aspects of open disclosure II: attitudes of health professionals — findings from a national survey. Med J Aust 2010. In press.
- 5. Studdert DM, Mello MM, Gawande AA, et al. Disclosure of medical injury to patients: an improbable risk management strategy. Health Aff (Millwood) 2007; 26: 215-226.
- 6. Kraman SS, Hamm G. Risk management: extreme honesty may be the best policy. Ann Intern Med 1999; 131: 963-967.
- 7. Boothman RC, Blackwell AC, Campbell DA Jr, et al. A better approach to medical malpractice claims? The University of Michigan experience. J Health Life Sci Law 2009; 2: 125-159.
- 8. Quinn RE, Eichler MC. The 3Rs program: the Colorado experience. Clin Obstet Gynecol 2008; 51: 709-718.
- 9. Kachalia A, Shojania KG, Hofer TP, et al. Does full disclosure of medical errors affect malpractice liability? The jury is still out. Jt Comm J Qual Saf 2003; 29: 503-511.
- 10. Iedema RAM, Mallock NA, Sorenson RJ, et al. The National Open Disclosure Pilot: evaluation of a policy implementation initiative. Med J Aust 2008; 188: 397-400. <MJA full text>
- 11. Australian Commission on Safety and Quality in Health Care. Open Disclosure Standard: a national standard for open communication in public and private hospitals, following an adverse event in health care. Canberra: ACSQHC, 2003. http://www.safetyandquality.gov.au/internet/safety/publishing.nsf/Content/compub-OD-Standard-2008 (accessed Feb 2009).
- 12. Health Services (Quality Improvement) Act 1994 (WA), Part 2.
- 13. Health Care Act 2008 (SA), Part 8.
- 14. Mental Health and Related Services Act 1998 (NT), ss 145-149.
- 15. Wu AW, Lipshutz AKM, Pronovost PJ. Effectiveness and efficiency of root cause analysis in medicine. JAMA 2008; 299: 685-687.
- 16. Health Services Act 1991 (Qld), ss 38G-38ZL.
- 17. Health Care Act 2008 (SA), ss 68-75.
- 18. Health Insurance Act 1973 (Cwlth), s 124Y(1), s 124Y(2).
- 19. Health Services (Quality Improvement) Act 1994 (WA), s 10.
Abstract
Health professionals worry that information about adverse events conveyed to patients in open disclosure (OD) may be used against them in medicolegal proceedings.
Whether and how strongly state and federal laws in Australia protect against such uses is unclear.
Our analysis concludes that existing laws do not prohibit the sharing of most types of information on adverse events with patients. However, none of these laws was enacted with OD in mind and, in general, the protections they provide are quite weak.
If policymakers want OD to become a routine part of medical practice, law reform may be needed in the form of stronger protections directed specifically at the contents of OD communications.