Gross negligence ‘medical’ manslaughter in Ireland: Legal context and clinician concerns

In recent years, the prospect of the criminal prosecution of medical practitioners for patient safety incidents resulting in fatality which occur in the course of clinical practice has caused heightened anxiety among medical practitioners, particularly in England and Wales, where a number of high-profile cases have raised public consciousness of this issue. The full impact of this landscape on individual practitioners and the delivery of healthcare has yet to be ascertained, although research suggests that medical practice has been impacted. Of particular interest is the phenomenon of defensive medicine which occurs where physicians adopt assurance and/or avoidance behaviours in an attempt to minimise the risk of medical negligence litigation and/or to avoid complaints to regulatory bodies. While defensive medicine is traditionally conceived of in a civil context, the possibility of criminal prosecution for patient safety incidents resulting in fatality may also result in alterations to medical practice. Drawing on the findings of an empirical study (a survey), this research sought to explore the impact, if any, of the threat of criminal prosecution on surgical practice in the Republic of Ireland, including a potential rise in defensive practice.


Introduction
In Ireland, redress for patient safety incidents resulting in harm is typically achieved through civil law remedies, that is, medical negligence litigation. Regulatory processes, such as complaints made to the Medical Council, 1 are also a feature of the landscape. In contrast, criminal prosecution for patient safety incidents have been rare and prosecutions of medical practitioners for gross negligence manslaughter, relating to their professional roles, have to-date been non-existent. Recent events, such as the CervicalCheck controversy, resulting from a failure to and/or delay in disclosing the results of a retrospective audit to a group of women, who had developed cervical cancer, 2 have increased the appetite for the accountability of the medical profession in Ireland. There has also been a criminal prosecution of a GP, Bernadette Scully, for gross negligence manslaughter in 2016, where it was alleged that the administration of an excessive quantity of sedative given during an epileptic seizure resulted in the death of her daughter. 3 While Dr. Scully was acting in a private capacity when the incident occurred, this prosecution has raised the possibility of prosecution in the consciousness of the public and within the profession. Increases in prosecutions of this nature in England and Wales, 4 are also significant given the close historic and cultural ties between the two jurisdictions and the similarities between the two legal systems. A factor which is poorly understood in this discourse is the lower legal standard which applies in gross negligence manslaughter cases in Ireland and which has the potential to lead to higher rates of conviction than in England and Wales, should the decision be made by prosecuting authorities to bring medical practitioners to trial for gross negligence manslaughter in Ireland. 5 family members "of a person whose death was directly caused by a criminal offence and who have suffered harm as a result of that person's death." Victims are now more powerful actors within the system having rights including a right to have a complaint acknowledged, a right to be informed about a decision not to prosecute and a right to have such decision reviewed. While it remains to be seen whether this changing landscape will impact directly on the prosecution of doctors for patient safety incidents, utilisation of these rights is likely to be a tool relied upon by grieving family members in some instances, thereby altering the investigation and prosecution process of patient safety incidents. The personal and professional impact on individual doctors practicing in an increasingly adversarial, high-risk environment, which includes the possibility of criminal sanction must not be underestimated. As Brazier and Cave note, '[t]he increasingly common demand that doctors be punished for tragic mistakes causes anxiety within the profession'. 6 Recent developments in the United Kingdom, such as the high-profile prosecution of Dr Hadiza Bawa-Garba, a junior doctor who had recently returned from maternity leave, for the gross negligence manslaughter of a young boy who tragically died under her care following a misdiagnosis, 7 has led to what has been described as 'toxic fear'. 8 Despite the systemic failures in this case including issues with the hospital computer system, staff shortages, and long-working hours, she was convicted in 2015, 9 and subsequently erased from the medical register, 10 leading to an outcry by the profession. 11 While the sanction of erasure was subsequently reversed, the case had a far-reaching impact. The implications of operating in this environment are felt by patients, doctors and the healthcare system, leading to defensive practice whereby the decisions of healthcare professionals are coloured by the threat of legal and/or regulatory action. The literature on defensive medicine has to-date been limited in two ways: it has focused on the impact on the patient and the system, to the exclusion of the physician; and has explored the impact of civil legal actions and/or regulatory sanctions, rather than criminal prosecution. Defensive medicine, the practices which may be adopted by physicians as a means of mitigating against the threat of litigation, complaints, and/or criticism, 13 arguably, could equally apply to measures taken to avoid criminal prosecution. While negligence litigation was traditionally viewed as the sole trigger for defensive practice, Case has described this view as 'reductionist' and notes that '[r]esearch into defensive practice in medicine has. . . re-contextualised such defensiveness as a multifactorial phenomenon with broader parameters'. 14 The acknowledgement that defensive practice is engaged in for a myriad of factors including 'for the purpose of protecting the doctor from criticism' 15 or 'protection against possible accusations of negligence or under performance' 16 is welcome, yet acknowledgments of the impact of potential criminal prosecution in this context have been rare. As Quick surmises 'whether the threat of criminal prosecution impacts on the behaviour of healthcare professionals, and if so in what ways, is not well understood'. 17 However, it appears from anecdotal evidence that prosecutions of this nature are resulting in defensive practice. For example, the independent review on gross negligence manslaughter commissioned by the GMC in 2019 noted: '[w]e have heard repeated reference to doctors resorting to defensive medicine'. 18 This article seeks to explore the impact of the threat of criminal prosecution on surgical practice in Ireland, drawing on the findings of an empirical study. The literature and case law suggest that there is an international awareness of the impact of the threat of litigation, criticism, and/or complaints on surgical practice, specifically, the area of practice which is the focus in this study. 19 Internationally, surgeons have been identified as a group who may be especially susceptible to defensive practices, due to the high-risk nature of surgery. 20  context of surgical practice are non-existent in Ireland. In addition, despite the significant body of research that exists in this area, research on the impact of the threat of criminal prosecution on the practice of medicine is limited. The first part of this paper critically examines the current legal framework as it relates to gross negligence manslaughter in Ireland, and discusses proposals for reform. The second part of this paper provides an overview of the methodological approach adopted in this study, and presents and explores the findings of this research. In doing so, the paper argues that an increase in prosecutions for gross negligence manslaughter has the potential to intensify defensive practice, and such a punitive approach to patient safety incidents which result in fatality will do little to cultivate a 'just culture' and achieve the overall important policy goal of patient safety.

Gross negligence manslaughter: the Irish position
Prosecutions for gross negligence manslaughter in Ireland, in a general context, are sparse, and non-existent in a medical context. The rationale for this approach has been underexplored in an Irish context. The Law Reform Commission, however, suggest that the low rate of prosecutions in this jurisdiction may be attributed to 'the reluctance of prosecutors to invoke the criminal law to deal with the negligent or incompetent discharge of lawful acts'. 21 The low number of prosecutions notwithstanding, it is necessary to provide an overview of the current legal framework, as medical practitioners in Ireland, are in theory at least, more vulnerable to conviction should a prosecution be raised than their counterparts in England and Wales, due to the lower legal standard which applies in this jurisdiction.
Gross negligence manslaughter is a form of involuntary manslaughter. There is no statutory basis for the offence, rather, it is a common law offence. The seminal test for gross negligence manslaughter in Ireland was established in The People (Attorney General) v Dunleavy, 22 where the death of a cyclist was caused by a taxi driver driving an unlit car on the wrong side of the road. Here, the Court of Criminal Appeal, in holding that the direction given by the trial judge to the jury was inadequate, 23 laid out a fourpoint test for the offence to apply: i. The accused was negligent by ordinary objective standards; 24 ii. That the negligence was of a very high degree; equal specialist or general status and skill would be guilty of if acting with ordinary care'.  iii. The negligence involved a high degree of risk or likelihood of substantial personal injury to others; and iv. The negligence caused the death of the victim.
Although Dunleavy involved a motor accident, the test applies to all incidents of gross negligence manslaughter and has been applied in the small number of cases which have been prosecuted since in this jurisdiction.
While the test as articulated in Dunleavy is similar to the approach taken in England and Wales, as established in the case of R v Adomako, 25 there is one significant difference. The Dunleavy test requires that the risk must be one of 'substantial personal injury', rather than risk of death. In contrast, the test for gross negligence manslaughter as enunciated in Adomako requires that the actions of the accused must have raised a risk of death, a risk of personal injury will not suffice. This standard has been confirmed by the English courts a number of times. For example, in R v Singh, 26 the Court of Appeal upheld the direction to the jury that 'the circumstances must be such that a reasonably prudent person would have foreseen the serious and obvious risk not merely of injury or even of serious injury but of death'. Similarly, in R v Misra and Srivastava, 27 the Court of Appeal confirmed that the risk is one of death not serious injury. Furthermore, in R v Rudling, 28 the court observed that 'a recognizable risk of something serious is not the same as a recognizable risk of death', which was subsequently cited with approval in R v Rose. 29 The Irish Superior Courts, however, have been consistent in their application of Dunleavy, and the lower risk required. The Dunleavy test has been affirmed and applied in the small number of cases which have been prosecuted since in this jurisdiction, including the death of a young woman following a chairoplane accident at a funfair, 30 and the death of a woman by neglect. 31 The lower standard as established in Dunleavy means, in theory at least, medical practitioners in Ireland are more vulnerable to criminal prosecution than their colleagues overseas. This is problematic, as it has been widely recognized that due to the inherent nature of risk in medical and surgical practice, doctors face greater potential exposure to the criminal law. 32  usually refuse to operate; risk (even risk of death) is an inherent part of medicine. Judgements have to be made instantly. The risk adverse doctor may do more harm than good'. 33 Criticisms have also been directed at the objective nature of the test. Like the test established in Adomako, the test as articulated in Dunleavy is objective and requires no subjective wrongdoing. This is particularly problematic in a medical context, given the high-risk nature of the practice of medicine. As Griffiths and Sanders explain, '[c]ases involving a momentary (but major) error with no evidence of recklessness or disregard. . . have therefore resulted in conviction'. 34 Much critique has also been levelled at the lack of clarity as to what may constitute 'gross' negligence. 35 In Ireland, the case law that defines 'gross' is of limited assistance, but it is necessary for present purposes to analyse the existing jurisprudence. In Dunleavy, Davitt J stated, If the negligence proved is of a very high degree and of such a character that any reasonable driver, endowed with ordinary road sense and in full possession of his facilities, would realise, if he thought at all, that by driving in the manner which occasioned the fatality he was, without lawful excuse, incurring in a high degree, the risk of causing substantial person injury to others, the crime of manslaughter appears clearly to be established. 36 In The People (DPP) v Cullagh, a case involving the death of a woman following an accident on a chairoplane funfair ride, the court described the degree of negligence necessary for a finding of gross negligence manslaughter as '. . . that degree of negligence which is appropriate to sustain a charge of manslaughter'. 37 Such a description easily lends itself to allegations of circularity. Indeed, the circular nature of the test for gross negligence was recognized in England by Lord Mackay in Adomako. 38 Such descriptions speak to claims that the test for gross negligence manslaughter is 'vague'. 39 Despite the limited guidance as to what constitutes 'gross' negligence in Ireland, in DPP v Joel, 40 a case involving the alleged neglect of an MS sufferer by her daughter and her daughters' partner, the High Court rejected a claim that the offence was unconstitutionally vague: 41. Ibid. 42. Evelyn Joel and Jonathan Costen were convicted upon re-trial and given a suspended sentence on the condition that they complete 240 hours of community service. This was subsequently overturned on appeal on procedural grounds. For a full discussion see, D. 'It is said that this kind of manslaughter is unconstitutional because it is vague. I do not agree. Criminally negligent manslaughter arises where the death of another person is caused in circumstances which objectively amount to a very high degree of negligence and which, in the circumstances in question, to any reasonable person the fact that a serious risk was unjustifiably taken with the life of another would be apparent. That emerges clearly and not as a matter of vagueness.'.. 41 Interesting, however, are some of the difficulties which were encountered in the Joel case, such as the failure of the jury to reach a verdict following a 7-week trial, which speaks to the inherent circularity of the offence. 42 Similar concerns around the vagueness of the offence have also been raised in England and Wales. In R v Misra and Srivastava, 43 two doctors argued that the offence was incompatible with Articles 6 and 7 of the European Convention on Human Rights, due to its uncertainty. However, these arguments were ultimately rejected. 44 Despite the judicial rejection of the contention that the offence of gross negligence manslaughter is vague in both Ireland, and England and Wales, a number of scholars have argued that 'gross' is in fact a nebulous concept. For example, Quick has asserted that gross negligence manslaughter 'remains an unduly vague concept that is incapable of objective measurement and consistent interpretation and thus potentially unfair to those prosecuted'. 45 Similarly, Griffiths and Sanders have described gross negligence as an 'intrinsically elusive concept'. 46 Such ambiguity also means, as Farrell et al. argue, juries are likely to encounter difficulties 'in determining whether a particular episode of substandard care on the part of a healthcare professional should be judged as criminal'. 47 The difficulties outlined with the objective nature of the test for this type of offence, coupled with the challenges often encountered in determining when an act or omission 48 may be considered 'gross' make this offence problematic. Several commentators have also highlighted practical problems in the context of prosecutorial decision making. Quick, in his study of prosecutorial decision making for example, evidenced prosecutorial unease 'with the brutality of bringing the full force of criminal prosecution to bear on individuals whose errors have catastrophic consequences'. 48 Wheeler et al. have also argued that due to the uncertainty as to what constitutes 'gross', 'those assessing whether to prosecute the crime are equally bewildered unable to calculate the likelihood of conviction'. 49 In an Irish context, the Law Reform Commission, commenting on the paucity of prosecutions in this jurisdiction, have also alluded to this concern, suggesting that the low number of prosecutions may be attributed to 'the reluctance of prosecutors to invoke the criminal law to deal with the negligent or incompetent discharge of lawful acts'. 50 Recent events, such as the Bawa-Garba case, have resulted in reviews of the law in this area and the recent recommendation of the Williams Review to develop an 'agreed and clear understanding of the law on gross negligence manslaughter' may therefore assist in ameliorating some of the uncertainty surrounding the offence in England. 51 In Ireland, a similar review and the development of prosecutorial guidance in relation to gross negligence manslaughter would also be welcomed. 52

Reform of the Irish position
Reform of the Irish position was considered by the Law Reform Commission (LRC) in its 2008 Report, which reviewed the law on murder and involuntary manslaughter as a whole. 53 In its review of the offence of gross negligence manslaughter the LRC considered raising the risk from 'substantial personal injury' to a risk of death, in line with the position in England and Wales, and also considered whether the offence should be abolished and replaced with an alternative such as negligent homicide.
Amending the risk as established by the Dunleavy test to that of a risk of death only, would be 'a modest but effective reform to the Irish position, and would provide medical practitioners some limited protection'. 54 However, the LRC recommended against a reform of this nature, noting that to raise the risk to one of death only would be 'too restrictive', and if it were to be amended, the risk should apply to both death or serious 55 55 Unfortunately, the LRC did not provide more detailed reasons for its reaching such a conclusion. The LRC also considered abolishing the offence of gross negligence manslaughter, and replacing it with a new lesser offence such as 'negligent homicide'. Although the LRC recognized that 'inadvertent killings are less culpable than intentional or subjectively reckless ones', 56 it ultimately did not recommend the replacement of gross negligence manslaughter. The LRC made few recommendations overall, noting that 'the law in this area generally functions well in practice'. 57 Thus, despite the criticisms discussed above, it appears that in Ireland, in the context of individual liability, the law in this area is settled for now.
A move towards organisational liability?   64 the Maidstone and Turnbridge Wells Trust was prosecuted for corporate manslaughter for the death of a patient, who suffered a cardiac arrest, following complications after delivery of her child by caesarean section. The case illustrated the need for sufficient evidence, as well as the high threshold for the offence, as ultimately, the court found no evidence of a gross breach and dismissed the case, noting 'it would be unsafe, and unfair to everyone. . . to leave this case to the jury'. 65 More recently, the Shrewsbury and Telford Hospital NHS Trust maternity scandal has raised public consciousness on the potential of corporate responsibility for wrongdoing. 66 The investigation and review into maternity care, which is ongoing at the time of writing, found an unusually high number of infant and maternal deaths and injury. 67 Given the low number of prosecutions for corporate manslaughter, it appears that the impact of the introduction of the statutory offence in England and Wales has been limited. While a detailed analysis of the 2007 Act is outside the scope of this article, it is useful to briefly consider some of the criticisms of the legislation and the possible reasons for the low number of prosecutions in this context. Kazarian, for example, has alluded to the inability of the offence to relate to healthcare institutions, noting that it is often difficult to prove causation in this context. 68 In addition, Roper has observed that there are significant costs involved in corporate manslaughter cases, which may partially explain why more prosecutions have not been brought. 69 This, coupled with the difficulty in establishing causation 'is likely to deter the CPS from tackling complex cases'. 70 The low number of prosecutions under the legislation and the shortcomings of the framework notwithstanding, Field has argued that 'the Act does have an important symbolic effect'. 71 Kazarian has also noted that although some of the traditional purposes of the criminal law such as deterrence often have a limited application to publicly funded healthcare institutions, other goals such as prevention of harm may be assisted by the 2007 Act. 72 In Ireland, reviews of the approach to corporate wrongdoing have been ongoing for the past number of decades. While there is a basis in the common law for such prosecutions, 73 the law in this area is muddied, and as Carolan notes, Irish law has not 'adopted a single, stable approach to the crime of corporate manslaughter'. 74 In 2005, the Irish Law Reform Commission recommended the introduction of legislation for corporate manslaughter, noting that 'as the current law of corporate liability for manslaughter does not provide a clear basis for constructing liability, a new basis, contained in legislative form, is necessary'. 75 The LRC recommended that the mental element of corporate liability for homicide should be equivalent to the existing test as outlined in Dunleavy for gross negligence manslaughter. The draft Corporate Manslaughter Bill 2016, if enacted, will create the statutory offences of 'corporate manslaughter' 76 and 'grossly negligent management causing death by a high managerial agent of the undertaking'. 77 Section 2 of the Corporate Manslaughter Bill 2016 sets out the offence of corporate manslaughter, outlining that an undertaking causes death by gross negligence where: a duty of care was owed; there was a breach of that duty, the breach of duty 'involved a significant risk of death or serious personal harm', and the breach caused the death of the individual. In contrast to the Dunleavy test, which requires a risk of 'substantial personal injury', the risk as required by the Corporate Manslaughter Bill 2016 must be 'a significant risk of death or serious personal harm'. Justifications for the differing approaches in this context are not apparent. The draft legislation, however, has been amended several times since 2005, and is not currently on the legislative programme for Government. This is problematic, as although a common law basis for this offence exists, McGrath notes that a prosecution is unlikely until there is a statutory basis for the offence. 78 Thus, it is unlikely that a healthcare organization would, at present, face criminal charges for manslaughter. Having provided a contextual overview of the legal framework and proposals for reform in Ireland, we will now progress to outline the methodological approach adopted in this study. We will then go on to explore the findings of this research, and discuss how the delivery of healthcare is affected by its environment, including the legal framework.

Methodology
This research was conducted as part of a wider study examining the impact of civil, criminal, and regulatory responses to patient safety incidents on surgical practice in Ireland. While a body of research has investigated the phenomenon of defensive medicine, to-date, there have been no studies in an Irish context. Quantitative methods were employed, with an electronic survey utilized as the data gathering instrument. 79 The survey instrument was informed by the literature, piloted among a small number of surgeons, and revised prior to dissemination. The survey was then sent electronically to all surgeons registered with the Royal College of Surgeons, Ireland (1,846 surgeons). A total of 1,251 opened the invitation to partake in the research, and 157 surgeons completed the survey, a response rate of 12.5%. SPSS analysis was subsequently completed. Qualitative analyses of 103 participant free-text responses to the open question '[w]hat is your opinion on criminal sanction for gross negligence which results in the death of a patient?' was also undertaken. Following the analysis of both the quantitative and qualitative data a narrative was formed. This research received ethical approval from the Education and Health Sciences Research Ethics Committee, University of Limerick.
There are a number of limitations with this research, namely the low response rate and the self-reporting nature of the research. Moreover, it is inherently difficult to measure defensive practice due to the myriad of factors which may contribute to changes in behaviour(s). These limitations notwithstanding, these findings provide previously undocumented insights into the impact of the threat of prosecution for gross negligence manslaughter on surgical practice in Ireland.

Results
The findings of this research suggest that the direct experience of surgeons in Ireland of criminal investigation or prosecution of safety incidents resulting in patient fatality made in the course of medical practice is very limited, with only two respondents reporting that they had been the subject of a criminal investigation in relation to their clinical treatment of a patient. That a very small number of surgeons had direct experience is unsurprising given the paucity of prosecutions for gross negligence manslaughter in this jurisdiction. Despite this, and the general absence of prosecutions for this offence in Ireland, 44 (33%) respondents reported that the threat of criminal prosecution for gross negligence manslaughter impacted on their medical practice currently, evidence for the first time of this phenomenon in Ireland.
Given the limitations of the quantitative data, the qualitative findings were more instructive, as they provided insights into the impact of the threat of prosecution for a patient safety incident which results in death, on surgical practice. Participants provided diverse views on the criminal prosecution of patient safety incidents which occur in the course of medical practice resulting in patient fatality.
Through an analysis of the research findings, three key themes were identified: (1) perceptions of prosecution for gross negligence manslaughter and the standard for liability; (2) the role of systemic factors; and (3) the impact on surgical practice and implications for healthcare delivery. These will now be presented, along with a discussion of the literature which is relevant to the research findings.

Perceptions of prosecution for gross negligence manslaughter and the standard for liability
In response to the free text query, the majority of participants indicated that they were clearly against the imposition of criminal sanctions in these circumstances, describing criminal sanction variously as: 'unfair', 'inappropriate', 'unwarranted', 'ridiculous', 'too harsh', 'unbalanced', 'counter-productive', 'pathetic', and 'problematic'. One participant noted that for the doctor who has made a genuine attempt to care for a patient despite poor judgement to have subsequent criminal proceedings is a terrible vista.
Surgeons also noted significant anxiety around criminal liability, describing the threat of prosecution as: 'horrifying', 'terrifying', 'frightful', 'an increasing worry' and 'extremely worrying'. One participant stated that the prospect of prosecution was Interestingly, there appeared to be some confusion about the current legal position in Ireland with a minority of participants clearly unaware that prosecution for gross negligence manslaughter is possible under Irish law at present. For example, the use of the phrase 'if it was brought here' in the passage above suggests a lack of awareness that gross negligence manslaughter is currently a criminal offence which physicians could face prosecution for in Ireland. Similarly, one participant asserted: '[it] is not on the Statute books in Ireland yet. . . no similar law is currently in force in Ireland'. This is despite the fact that, while there has been a dearth of prosecutions for gross negligence manslaughter in a general context, and as previously outlined, none in a medical context, due to the lower standard that exists in this jurisdiction, a prosecution of this nature would, in theory at least, be easier to secure.
Perhaps surprisingly, 36 participants accepted that prosecution for gross negligence resulting in death was appropriate in some circumstances. It was suggested that it '[s] hould be extremely rare' and '[c]ritically depends on situation and context'. Respondents indicated that responses should be proportionate and investigated thoroughly and fairly: 'an investigation is appropriate; a criminal sanction can only be made if proven significant'. Interestingly, this cohort of participants spoke of criminal sanction being appropriate if the actions of the doctor were 'wilful', with 9 participants suggesting that criminal sanction was appropriate only if wilful harm/death/negligence involved. While it is difficult to ascertain what this meant to respondents, it is likely associated with a form of intention/intentional act.
These comments highlight some of the difficulties discussed above with the current approach to the prosecution of doctors for safety incidents resulting in the death of a patient. The standard of culpability is one of gross negligence rather than an intent to cause harm or recklessness, and the test is an objective one, with the result that a doctor could be convicted of gross negligence manslaughter in respect of an action where there was no intention to harm, the doctor did not act recklessly, or indeed, in a situation where there was no subjective wrongdoing. For this reason, scholars such as Quick have previously argued that the offence should only be invoked where conduct is reckless. 80 These arguments are surely at their strongest when made in contemplation of a profession where risk to patient life is often an inherent part of everyday practice and practitioners take on this responsibility despite the potential personal and professional cost. In the independent review of the law on gross negligence manslaughter conducted at the request of the GMC in response to the outcry within the profession to the case of Dr. Bawa-Garba, it was noted that although outside the terms of reference of the review: it is right to record that many of those who provided evidence to the review thought the law should be changed. They argued that for an act or omission to constitute a criminal offence it should involve either a deliberate act leading to harm or reckless indifference to the consequences of an action. 81 Interestingly, this response was not limited to the medical profession and its representatives. The issue of whether it was appropriate to criminalise patient safety incidents, even those considered to be exceptionally bad, was described as 'the most fundamental issue raised' during the public consultation. In both written and oral submissions, the 'very clear view' was that a mistake should not be sufficient for criminal liability, '[t]he doctor's actions must have been intentional or reckless and the outcome for the patient resulted in lasting harm or death. Any attempt by a doctor to cover up, falsify or blame others for clinical errors also implied criminality '. 82 Participants' also demonstrated concern with the lack of consideration given to systemic issues including an overstretched healthcare system with attendant staff shortages which contribute to patient safety incidents that occur daily, within these processes.

The role of systemic factors
Like some other high-risk professions, it is generally accepted that medical professionals are more exposed to criminal sanction given the nature of the work in which they are engaged. 83 This was also highlighted by participants, with one surmising: My opinion is that is what is frequently lost in commentary on cases such as these in the media as well as the medical landscape, is that events such as these are extraordinarily rare, and are a result, at the end of the day, of the efforts of one human being (i.e. the doctor) to help another. There is no other motivating factor. Unfortunately, in our line of work complications and mistakes can result in [an] adverse outcome, even death, and reflect the enormous responsibility that we carry.
This speaks to the oft-cited issue that consideration is not given to the circumstances surrounding the incident. 84 In this research, a number of respondents pointed to the systemic and resourcing issues which contribute to an environment where patient safety incidents occur, sometimes catastrophic ones: [i]t is very worrying in the day-to-day management of very sick patients with an under resourced health service and shortage of high-quality junior doctors as frontline staff.
Another elaborated on this and pointed to the need for a: balance [sic] between a measure for standard of care and other factors which caused the medical negligence in first place. Usually this is due to doctors working in high demand setup with minimal time allocated, multitasking and burnout factors, after all doctors are humans too.
It was clear that surgeons were dissatisfied with potentially being held criminally responsible for actions taken in an overburdened system: if the threat of criminal sanction is to be applied. . . then adequate resources have to be in place [sic] for delivery of care in terms of staffing and resources and moreover, acceptable and safe levels of care codified. Otherwise such a sanction is untenable and makes many potentially high-risk cases untreatable.
Participants also argued that the medical profession is vulnerable in the face of the legal system which 'looks at outcome rather than process'. The failure of the legal system, and the criminal justice system, to take systemic issues such as the understaffing of hospitals into account is a contentious issue among the medical community, with the prosecution of Dr Bawa-Garba bringing many of these concerns under the spotlight. 85 These systemic issues have also been highlighted by a number of scholars, including Samanta et al., who have argued: [f]rom an ethical perspective, it is difficult to understand why the focus should fall entirely on the individual (unless there is specifically egregious conduct on the doctor's part) in cases of death due to gross negligence that occurs within organisations where legitimate clinical care is delivered. Surely there is an organisational responsibility to provide a safe environment with adequate staffing and appropriate safety netting. Yet systemic factors seem to carry low weight in the general response to such cases. 86 Such arguments suggest that a move from individual to organisational liability would be preferable, and a better means of ensuring accountability. As previously discussed, manslaughter prosecutions against corporations may be brought in England and Wales under the Corporate Manslaughter and Corporate Homicide Act 2007. In Ireland, there is a common law basis for such prosecutions, with reform in this area pending under the guise of the Corporate Manslaughter Bill 2016. Despite these reforms, prosecutions against healthcare organizations have to-date been scant, with commentators arguing in the context of England and Wales, that the legislation is complex, 87 and in reality, 'it is easier to convict an individual doctor for gross negligence manslaughter than it is to effect similar accountability on an organisation'. 88

Impact on surgical practice and implications for healthcare delivery
While recognising that 'impact' is an inherently difficult thing to measure, and the primary limitation of most empirical studies on defensive medicine is the self-reporting nature of the research, 89 the findings of this research are instructive in that they highlight significant concerns among Irish surgeons in relation to the prospect of criminal prosecution and its potential impact on both individual doctors and the delivery of healthcare in Ireland, suggesting that defensive practices exist in this jurisdiction for the first time. 90 Defensive medicine has been described as 'a doctor's deviation from development of medical science and the supreme importance of that development to humanity makes it particularly undesirable and inconsistent with the common good that doctors should be obliged to carry out their professional duties under frequent threat of unsustainable legal claims.' Defensive practice has also been identified as a policy concern in jurisprudence in England and Wales. Case, for example, has reported that 'there are at least 49 reported cases where defensive practice arguments are expressly referenced.' Case notes that these were most frequently associated with police liability cases. See, Case, 'The Jaded Cliché of "Defensive Medical Practice"', p. standard practice to reduce or prevent complaints or criticism'. 91 Divided into two strands, defensive practices may include 'assurance' or 'avoidance' behaviours. The former including actions such as clinically unnecessary referrals for additional tests and/or second opinions, and the latter encompassing practices such as refusing to engage with high-risk patients. 92 Both 'assurance' and 'avoidance' behaviours have been recognized as problematic for a number of reasons. It has been suggested, for example, that clinically unnecessary referrals contribute to heightened healthcare costs and limit healthcare resources. 93 Additional referrals may also expose a patient to further risks. 94 Avoidance behaviours are also particularly problematic from a patient safety perspective, as these behaviours can result in the 'avoidance of certain patients and procedures, thereby withdrawing medical services and can deny patients productive care'. 95 Respondents to this study, pointed to the potential for avoidance behaviours among surgeons in Ireland: '[d]octors are likely to avoid high risk specialities because of this' and that it 'tends to lead to defensive practice of medicine in Ireland'. Another respondent highlighted the steps, s/he would take and the consequent implications for patient care: [i]f it became the norm I would take significant steps to reduce my practice and its risk profile (e.g. would stop offering the second opinion and complex surgical options I currently provide).
These findings are similar to other empirical studies in this area which have focused more broadly on the impact of civil and regulatory frameworks on clinical practice, and supports the idea that doctors will take steps to avoid high risk patients or procedures in order to minimize their risk profile, 96 a motivation which is likely to intensify when a potential criminal conviction may result.
In the context of this research, respondents spoke to the impact of the threat of criminal prosecution on the surgical profession, including recruitment, with one respondent observing that The threat to the future of the profession, whether by virtue of emigration or early retirement, is an issue within the defensive medicine literature more generally. For example, Nash et al. have commented on this issue, noting 'accelerated retirement may contribute to workforce problems at a time when most medical disciplines have national shortages'. 97 It was suggested one respondent to our study that '[t]he Irish health service is currently heading to a situation whereby it will become impossible to practice'. Ultimately, catastrophic consequences were predicted: The idea of criminal sanction in circumstances where there is no pre-meditated determination to harm is abhorrent, and combined with the toxic medicolegal environment in this country, would signal the end of medical practice as we know it. It would cause an earthquake, and ultimately the ones to suffer the most would be patients awaiting treatment.
Another interesting contribution related to the impact of potential prosecution for gross negligence manslaughter on practices following a patient safety incident in a system where policy, and more recently, legal measures encourage open disclosure and reflection: [It is a] very unbalanced, unfair system and clearly counter-productive when open disclosure is mandated by the employer and then all genuinely disclosed facts can be used to criminally prosecute the individual doctor with no accountability for the institution or systems involved. The current system does not encourage learning from our mistakes, rather it makes doctors less likely to be open in fear of prosecution. This is wrong! Quick has highlighted concern about the impact of criminal prosecution on a culture of candour, noting '[p]rosecuting professionals for negligence . . . can be seen as a somewhat draconian response that risks increasing secrecy rather than openness about safety'. 98 This concern may come under further scrutiny in Ireland when the Patient Safety (Notifiable Patient Safety Incidents) Bill 2019 is enacted. 99 The Bill will introduce the mandatory open disclosure of 'notifiable patient safety incidents'. 100 Similar to the provisions of the Civil Liability (Amendment) Act 2017, which provides protection to voluntary open disclosures made following patient safety incidents, 101 the proposed legislation will mandate the disclosure of serious patient safety incidents and prevent such disclosures from being admissible in civil legal proceedings, regulatory complaints processes, and from invalidating insurance. 102 However, it appears that this legislation is framed without contemplation of the criminal context. 103 The second, related, issue raised in our research relates to the value of reflective practice in modern medicine and the impact of a culture where practitioners fear criminal prosecution, particularly the use of reflections engaged in as part of a self-improvement process, as evidence in a criminal trial. Mullock has noted that an increase in prosecutions for gross negligence manslaughter may encourage a climate of secrecy 'that obstructs learning from error and improving patient safety'. 104 This issue has been the subject of contentious debate in England and Wales recently in the case of Dr. Bawa-Garba, where a mistaken belief arose among many within the medical profession that reflections produced by Dr. Bawa-Garba were admitted in evidence and used against her during the course of the trial. 105 This was again highlighted in the GMC initiated independent review on gross negligence manslaughter: '[w]e have heard repeated reference to doctors resorting to defensive medicine and refusing to engage in learning and reflection following incidents for fear that this could be used in evidence against them'. 106 While Bawa-Garba's learning reflections were not part of direct evidence in her trial, it is possible as Samanta et al. suggest, that 'there may have been indirect use of these statements'. 107 It is clear that the criminal prosecution of doctors will have a lasting impact on the health system by reducing opportunities for reflection and learning and moving medical professionals back towards a culture of secrecy, impacting on patient care and the doctor patient relationship. Moreover, the findings of our research highlight significant concern among surgeons in Ireland. This is despite the fact that a physician is yet to face prosecution for gross negligence manslaughter in respect of their professional role. Thus, our findings suggest that the implications of such prosecutions on practice are likely to be significant in jurisdictions such as England and Wales where these prosecutions are more common.

Conclusion
To-date, prosecution for gross negligence 'medical' manslaughter has been non-existent in Ireland. Heightened societal demands for accountability of the medical profession, 108. Dr Bernadette Scully was prosecuted in 2016 in respect of the death of her profoundly disabled daughter from a drugs overdose. 'Bernadette Scully found not guilty of daughter's manslaughter', Irish Times, 16 December 2016, available at https://www.irishtimes.com/ news/crime-and-law/courts/circuit-court/bernadette-scully-found-not-guilty-of-daughter-smanslaughter (accessed 30 August 2020 coupled with an increase in prosecutions in England and Wales, suggest that recourse to the criminal law as a response to patient safety incidents resulting in fatality may become a more frequent part of the landscape in this jurisdiction.
Our findings suggest that there is considerable disquiet among surgeons about the prospect of prosecution of medical professionals for gross negligence manslaughter in the Republic of Ireland. This is despite the fact that there has yet to be a prosecution of a medical professional for gross negligence manslaughter in respect of their work as doctor. 108 However, healthcare professionals working in Ireland are actually more likely to be convicted should they be prosecuted given that the level of risk to which patients must be exposed is that of a 'high degree of risk or likelihood of substantial personal injury' 109 rather than an obvious and serious risk of death in England and Wales. 110 The lower standard applicable in Ireland is especially problematic for the healthcare profession given that the nature of the work in which they engage leaves them particularly vulnerable, a point noted by respondents to this survey.
The findings of this research suggest that although there are varying levels of awareness of the legal position in Ireland among surgeons, respondents who were aware of the possibility of criminal sanction in Ireland expressed real concern. The findings indicate that should awareness of the legal position in this jurisdiction be heightened, for example, through a prosecution of a physician for gross negligence manslaughter, there would be a significant impact on practitioners, practice, and the delivery of healthcare in Ireland.
These findings provide an insight into the experience and views of medical practitioners directly impacted by the threat of criminal prosecution for mistakes made in the course of their professional lives. Surgeons working in the Republic of Ireland report a range of previously undocumented impacts of this threat and the factors which contribute to the practice of defensive medicine in response to threats of criminal prosecution. The study expands our understanding of the impact and reach of defensive practice, focusing on criminal rather than civil sanction and the impact on the individual practitioner, in conjunction with more traditionally accepted impacts on the patient and the healthcare system.