BUSINESS MANAGEMENT GREAT WORK, ON TIME, NO PLAGARISM, A+ WORK

Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725 1 of 7

Conflicts of interest in clinical ethics consults Elliott Mark Weiss ,1,2 Aaron Wightman,1,2 Laura Webster,3,4 Douglas Diekema1,2

Original research

To cite: Weiss EM, Wightman A, Webster L, et al. J Med Ethics 2021;47:e61.

1Department of Pediatrics, University of Washington School of Medicine, Seattle, Washington, USA 2Treuman Katz Center for Pediatric Bioethics, Seattle Children’s Research Institute, Seattle, Washington, USA 3Department of Hospital Consultative Services, Virginia Mason Medical Center, Seattle, Washington, USA 4Department of Bioethics and Humanities, University of Washington School of Medicine, Seattle, WA, USA

Correspondence to Dr Elliott Mark Weiss, Pediatrics, University of Washington School of Medicine, Seattle, WA 98195, USA; emweiss@ uw. edu

Received 25 July 2020 Revised 9 October 2020 Accepted 15 November 2020 Published Online First 21 December 2020

© Author(s) (or their employer(s)) 2021. No commercial re- use. See rights and permissions. Published by BMJ.

ABSTRACT Although there is wide agreement that ethics consults are at risk for conflicts of interest (COIs), ethics consultants (ECs) have limited guidance with regard to how to identify and approach COIs. We aim to address these concerns and provide practical guidance. We will define and consider four categories of COIs: consult type, team composition, dual clinical roles and other concerns. We will define and consider six actions available for ECs to take in response to COIs: no action, disclosure only, obtaining a second opinion, referring to another EC, referring to an institutional ethics committee or seeking an outside consult. We will then propose a points- based algorithm for ECs to use to determine the appropriate response to COI. Finally, we will discuss the strengths and limitations of our proposed algorithm.

CASE #1: NEONATE WITH NEUROLOGICAL IMPAIRMENT AND KIDNEY DISEASE A neonate in the intensive care unit suffered profound neurological impairment from severe hypoxic ischaemic encephalopathy and remained dialysis dependent with end- stage kidney disease from the same ischaemic insult. During the patient’s hospital course, numerous areas of disagreement arose between the medical team and family when determining goals of care and considering decisions for life- sustaining treat- ments. Members of the medical team questioned the patient’s suitability for future kidney trans- plant and called the ethics consult service. At the time the consult was placed, the on- call ethics consultant (EC) had not yet met the family, but was familiar with the case because she was serving as on- call nephrology consultant responsible for the patient’s dialysis overnight concurrently with EC responsibilities. The EC’s two main areas of scholarship were decision- making for neonates with profound neurological impairment and for neonates with end- stage kidney disease. The EC was concerned about a possible conflict of interest (COI) and therefore spoke with a second EC. This second EC was also a neonatologist. He had recently participated in many contentious conver- sations with this patient’s family. The second EC was also concerned about COI if asked to serve as the EC for this case.

Although many would agree that ethics consults are at risk for COIs, ECs have limited guidance with regard to how to identify and approach COIs. We aim to address these concerns and provide practical guidance. We will examine the roles of consult type, team composition, dual roles and other concerns in formulating our guidance.

BACKGROUND Ethics consult services began nearly a half century ago as one of many responsibilities of hospital insti- tutional ethics committees (IECs).1 Today, ethics consult services are widely endorsed by profes- sional societies.2–4 While it may seem intrinsic to the endeavour of ethics consults that they must be impartial, neutrality in this context is a surprisingly complex phenomenon.5 6 Nonetheless, it is clearly problematic for the EC to be or be perceived to be beholden to parties actively involved in a consult. As Meyers describes, although it may be tempting to identify a particular client in specific consults (eg, this patient vs this physician) or generally (eg, patients vs hospital administrators), this is the wrong approach:

Ethics consultants are not, and cannot rightly be, agents; they do not act on behalf of any person or persons, but instead on behalf of doing the right thing.7

Meyers claims that, in this way, the role of the EC is unique within the hospital because rather than advocating for a particular agent, she must promote an ethically justifiable outcome or process. The American Society of Bioethics and Humanities (ASBH), the primary professional organisation in North America for clinical ECs, has produced and intermittently updates a Core Competencies docu- ment, in which they identify the core knowledge and competencies required of EC to ensure quality of ethics consults.4 Similarly to Meyers, the ASBH document suggest that ECs ought to advocate for better outcomes globally, rather than on behalf of a particular party:

The general goal of ethics consultation is to improve the quality of healthcare through the identification, analysis and resolution of ethical questions or concerns.4

These ideals overlook the reality that many ECs serve a dual role as clinicians and most are paid by the institutions within which they work, raising a concern for COI.8–10 COI may be inherent, to some degree, in all professional relationships. This fact does not diminish the importance of attempting to recognise and address it where it impacts patient care. The ASBH Core Competencies require consul- tants to be honest about how their own agenda and values may shape the consult, but do not provide clear guidance on how to identify a potential COI and what action, if any, a consultant should take.

There is a growing recognition of COIs in ethics consults.7 11 Spielman identifies three types of COIs for ECs: those stemming from relationships with clinicians; those stemming from commitments to particular bioethics positions; and those stemming

2 of 7 Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725

Original research

from institutional pressure to settle or resolve cases.12 Frolic and Chidwick have emphasised the difference between ‘in- role’ COI intrinsic to the EC role and ‘out- of- role’ COI such as those stem- ming from personally held values.13 In the paediatric context, Lyren and Ford note that consultants and committee members should be ‘self- reflective about the need to recuse themselves when an assessment may be inappropriately or unduly influ- enced by personal experience or dual role’.14

However, minimal guidance exists regarding how to prag- matically address these concerns. Even within the realm of rela- tively straightforward financial COIs, it can be difficult to know whether a COI exists and, if so, how best to move forward to address the relevant issues.15 The ASBH Core Competencies Update Task Force4 states that ECs should attempt to avoid COIs and, if they are identified, should be managed through ‘skilled disclosure, recusal or transfer to another consultant’. The task force does not give specific advice regarding how to identify a COI or how to determine which response to select. Similarly, the ASBH Code of Ethics encourages avoidance and ‘if it is not possible to avoid such a conflict, then it should be managed using ethically supportable strategies’.16 Together the ASBH documents support some level of commitment to iden- tifying and addressing COI, but strategies to do so are not well defined.

Before continuing, we will make two observations about wording choice and categorisations. First, some have made distinguishing a ‘possible COI’ from an ‘actual COI’, a central part of the ethical calculus. For example, the ASBH Code of Ethics states that ‘discerning between a possible conflict and an actual conflict requires vigilance and continual self- reflection’.16 We believe that this difference is difficult to defend rigorously. More importantly, the perception of COI (possible or actual) by third parties may impact the quality of ethics consults. There- fore, we will proceed without discriminating between the two. Second, some authors differentiate between COI and ‘conflict of obligation’.16 We feel that this distinction is not useful as both types of conflict are similarly approached. Here we define COI as competing interests or responsibilities that have the poten- tial (or appear to have the potential) to adversely influence or compromise the professional judgement of the EC, the process, and/or outcome of an ethics consult.

Note that here we use terminology for clinical positions that is standard in North America but may not be commonly used elsewhere. The attending physician is the senior responsible physician on a clinical service. A consultant is an individual who is asked by a patient’s primary attending physician to provide recommendations related to that person’s particular area of expertise (eg, bioethics). Similarly, there is regional and inter- national variation in who performs ethics consults. We include three potential parties: individual ECs, small groups of ECs and assembled IECs (known as clinical ethics committees in some countries). Internationally, the frequency of using each of these options varies, but that variation does not need to negatively impact the usefulness of our categorisations. Our tool is intended to apply individually to each EC involved in a case.

We propose a new model for identifying and addressing COIs proactively as they arise in clinical ethics consults. We start by considering four categories of COIs in table 1: (1) consult- type COI; (2) dual- role COI; (3) team- composition COI and (4) other COI. Next, we present six actions available to ECs to address or avoid COIs in table 2. This differentiation into types of COI and methods to manage them parallels how other experts have considered COI, such as the American Bar Association.17 We then offer a proposed algorithm to guide ECs as they consider

how to respond to COIs. Finally, we will review strengths and weaknesses of the model.

Consult-type COI In our clinical experience, the content of most ethics consults can be categorised into one of three types of increasing risk of COI becoming more problematic. First, those asking the consul- tant a simple question or to provide a framework for addressing a common issue; second, those focused on issues that have led to moral distress among some clinicians; and third, those in which there is a disagreement or dispute.

The first category of ethics consult type includes two subcate- gories. First, simple questions to which there is a straightforward, generally agreed- upon answer, such as how to determine whom the designated surrogate decision- maker is, whether a compe- tent 19- year- old can refuse life- saving surgery or clarification of institutional policy. Second, requests for providing a moral framework, such as ways to consider limits to parental authority or explanation of how to interpret a standard principle. In our clinical experience across multiple institutions, these consults

Table 1 Categories of items relevant to conflicts of interest for ethics consultants (ECs)

Category Key features Example

Consult type Provide moral framework or objective answer

Team asks for guidance implementing Harm Principle in a particular case

Moral distress Team is not asking for guidance on what to do, but wants support for bedside nurses reporting moral distress

Dispute Team asks for guidance in settling a dispute with upset parents

Dual role Past EC was previously patient’s primary nurse

Active EC is currently patient’s social worker

Future EC will become patient’s primary attending next week

Team composition

Committee More than 4 ECs consider case before recommendations

Small team Small team of ECs (2–4 individuals) serve together to make recommendations

Individual Single EC considers case and makes recommendations

Other Interpersonal Early career neonatologist/ethicist is EC for consult requested by NICU division chief

Idiosyncratic position EC has published articles promoting idiosyncratic positions related to case

Bias EC has a fixed conclusion based on genetic diagnosis before meeting with stakeholders

Financial EC asked whether to order a very expensive test on which he holds a patent

NICU, neonatal intensive care unit.

Table 2 Actions available to address or avoid conflicts of interest (COIs)

Degree of concern about COI Response Continue EC role?

Lowest Take no action Yes

Disclosure only Yes

Obtain a second opinion Yes

Refer case to a colleague No

Refer case to the committee No

Highest Seek outside consult No

EC, ethics consultant.

3 of 7Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725

Original research

require the least degree of interaction with clinicians, patients and families. Consults in which the EC interacts with the patient and family have the greatest potential for helping or hurting the patient.4 Therefore, consults without this interaction are at lowest risk for COI, and clinicians can generally feel comfortable serving in the EC role regardless of other issues.

The second category of ethics consult type focuses on issues around clinician moral distress. Examples include a nurse being asked to do something that may be causing suffering or a clini- cian who feels the patient or surrogate has made a choice that is unsafe or unreasonable. These types of consults are commonly encountered by ECs.9 They typically do not involve two key components that make COIs problematic: contact with patient and family, and the development of recommendations related to the particular patient.18 From quite early in the history of ethics consults, ECs have appreciated that making specific guidance regarding a particular patient creates challenges beyond creating or following guidelines more generally.19 Therefore, these are also at lower risk for COI. However, if the EC is intimately involved in a clinical role associated with the current case, then the clinician requesting guidance related to moral distress may feel that this individual is biased or may have contributed in some way to the conditions leading to moral distress. In these situations, greater action may be required.

Consults in the third category involve a disagreement or dispute between parties which may require informal negotiation or, less commonly, formal mediation (eg, between patient or/ and clinicians, between patient and family, between clinicians, or between patient, family, or clinicians and the institution). Examples include disagreement regarding end- of- life treatment, surrogate refusal of efficacious treatment and treatment plan/ surgical candidacy disagreements. This category also includes when stakeholders consult ethics because they feel uncertain about how to proceed. A foreseen or potential dispute can be among the most charged discussions and has the potential to lead to intractable conflict. Navigating between multiple options that all seem problematic or figuring out what to do when there may not seem to be any viable options available is meant to be categorised similarly as active disputes. Disputes are typically the most contentious cases encountered by ECs and are among the most common reasons for ethics consults.8 They require the highest degree of involvement from the consultant who must work closely with all impacted parties or to develop recommen- dations. As a result, this category of consult is at the highest risk of COI and therefore actions should be taken to avoid COIs.

Dual-role COI ECs may serve many additional roles at their institution. Surveys of ECs routinely show that clinicians make up the majority of ECs. In a national US survey, Fox et al found that nurses and physicians each made up one- third of ECs, with social workers, chaplains and hospital administrators each comprising 10%.8 Within paediatrics, clinicians made up the bulk of IEC members9 as well as members of the ethics consult services.10 In addition to roles as ECs, clinicians may also serve administrative roles in the hospital including supervising other clinicians or serving on Institutional Review Boards. While clinicians remain the focus of this paper, it is worth noting that non- clinician consultants can also have dual- role COIs. For example, a clergy member who serves as EC may also be asked to provide spiritual care or find that their primary responsibility as a spiritual guide within a particular religion might conflict with certain types of ethical concerns.14 For hospital administrators who serve as ECs, a conflict between the duty to protect the institution and duty

to protect the patient’s interest may exist.2 There may also be conflicts when ECs have other relationships (eg, spouse) with the requesting clinician.20

The dual roles of clinician and EC raise the possibility of COI if there is perception that the clinical role of the clinician has influenced the ethics consult, usually in favour of the medical team and at the expense of a patient or their family.4 12

The temporal relationship (past, current or future) between the consultant’s interaction as a clinician and as an EC is an important factor in the consideration of COI. Situations when the EC simultaneously serves as the patient’s clinician are the most concerning for COI. The particular details of the active clinical role may also be important. For example, relationships may be different if the EC acts as the primary attending, nurse, or heavily involved specialist consultant compared with the overnight attending or a consultant with limited interaction with the patient or their family and less active role in overall medical decision- making. Is it less problematic to be providing night- time call coverage than to be the daytime rounding attending physician? Is a ‘curbside’ ethics consult subject to the same level of scrutiny as a formal consult? The specifics of the role may influence the degree of COI present and guide the chosen action to remedy concerns. In most of these circumstances the likeli- hood of COI will be high enough that the EC should consider taking some action.

A past role as a clinician may also create a COI for an EC. These must be considered on a case- by- case basis. In many situations, an individual who had minimal interaction with a patient and family may reasonably serve as EC. This should be approached cautiously, however, as even previous interactions considered ‘neutral’ or without conflict by the clinician may have had a greater impact on the patient or family than perceived. More significant previous encounters (perceived by the EC, patient, family or medical team) could impact future ability to provide ethics consults. Past negative encounters, characterised by disagreement or distrust between the EC and patient/surro- gate or the primary clinical team, will make serving in the EC role difficult, if not impossible, and action should be taken to avoid assuming the EC role. Past positive encounters may also present a problem. On the one hand, a trusted individual serving as EC might aid in the ability to mediate conflict for a particular family. However, in other situations, it may cloud the judgement of the EC (eg, in favour of the patient) or may create role confu- sion for the patient and family.

Finally, a presumed future role may also create COI. For example, consider an oncologist currently serving the EC role, who receives a consult for an inpatient for whom they will become the primary attending the following week. Although the EC has no history with the patient and no current dual role, it may be reasonable to take action to avoid COI proactively in such situations.

Team-composition COI Ethics consult team composition is relevant when considering the possibility of COIs. In a national survey of ECs, Fox et al found that most consults were performed by a small team (68%), rather than by full committee (23%), or a single individual (9%).8 When a single individual performs the duties of ethics consult with minimal oversight, the potential risk of COI is highest. In most cases, the likelihood of conflict decreases if the EC is a member of a team that performs consults collaboratively and in accordance with standard practice. Likelihood of conflict falls further if the EC is participating in the ethics consult as a member of the full IEC with diverse perspectives. There may be situations

4 of 7 Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725

Original research

where an EC should not perform an ethics consult as an indi- vidual due to COI, but it may be permissible to serve as part of a full ethics committee for the same consult. While a larger team performing does not fully address a conflict, multiple perspec- tives provides a check compared with a single EC. In situations where the individual has significant COIs, the individual should consider recusing oneself from participation. When a team is being used, each individual EC should consider their own risk of COI. That allows for different scores for different potential members of a consult team or committee. The individual can then decide what actions, if any, are appropriate to take.

Other COI Interpersonal concerns can arise related to the relationship between the individual who is both a clinician and EC and their clinician colleagues.12 This desire to maintain positive relationships could bias ECs towards siding with a colleague rather than a patient and their family. An example could be a consult regarding disagreement over end- of- life treatments in a patient for whom the medical team feels significant distress about continued life- sustaining treatment. Concerns may also arise when working with a superior, in which case the pressure towards an outcome satisfying to the medical team is likely even stronger. For example, consider if a junior clinician serves as the EC for a case in which the primary attending is the division chief. Relationships between ECs and those requesting the consult can create COI in rather complicated ways.20

Interpersonal concerns involving particularly close relation- ships are a subset that merits mention. As others have high- lighted, the nature of the relationship matters: it is different to be married to a party in the consult than to be the mentee of a party in the consult.20 If an EC is, for example, a family member or sexual partner of the subject of the consult, recusal is clearly warranted. This is probably true for close friends and colleagues, but to a less degree for neighbours, coworkers or other acquain- tances. The nature of human relationships varies infinitely; each of these must be assessed individually to determine whether recusal is appropriate.

Particular ECs may be committed to particular idiosyncratic positions that influence their consult. Spielman gives the example in which an EC, who had written articles claiming that advance directives should sometimes be discarded, concluded (against state law) that the patient’s advance directive should be ignored and referred the consulting clinician to his own published article.12 Sometimes this may be more subtle, for example, if the consultant in the opening vignette was committed to dialysis and transplant provision for all neonates regardless of intellec- tual function, a view not shared by a majority of neonatologists, nephrologists or ethicists. Although these situations may seem uncommon, given the nature of ethics consults, contentious issues that may remain unsettled (even among experts) likely make up a large proportion of consults. We acknowledge that bioethical views evolve over time and it is possible that what is an idiosyncratic view today may be recognised as the preferred view in the future. Nonetheless, we feel that strongly held views that substantially stray from accepted bioethical principles and policies create potential for COI that should be considered. ECs must have a level of self- awareness: when an issue they feel passionately about is the focus of a clinical consult or to which they have an ideological commitment that might not be shared by the patient or family, taking action to address COI may be warranted.

All ECs carry biases that have the potential to influence their decisions.21 ECs are obligated to be aware of and respond

appropriately to their own personal and professional biases.16 Full attention to explicit and implicit biases and how they, know- ingly or unknowingly, influence decision- making and contribute to systematic inequities including racism is beyond the scope of this paper.22 Nonetheless, ECs should strive towards increased awareness of when their own biases may be influencing a consult, and bias should be seen as a potential to generate COI.

Finally, ECs may have financial relationships with the partici- pants of an ethics consult or may benefit financially due to clin- ical decisions.

Potential responses to COIs We differentiate six distinct actions available for ECs to address or avoid COIs (table 2) with more significant actions corre- sponding to the increasing level of concern raised by the COI. First, the EC may take no action: continue to serve as EC without any corrective action. Second, the EC may disclose the COI only. The ASBH Core Competencies Update Task Force promotes ‘skilled disclosure’, defined as ‘proactively clarifying expectations and clearly identifying when he or she is acting as an EC or in another role’.4 Skilled disclosure should include disclosing to all parties involved in the EC. Finally, it may be wise to disclose COI to those in the role of oversight, such as the IEC, risk management or to hospital administration. Proactively informing hospital administration of a COI may be good prac- tice, in particular when specific concerns are present that the situation may require escalation to the point where the EC may feel compelled later to no longer serve in the role.

Third, the EC may obtain a second opinion by presenting the case, the stakeholder viewpoints and the ethical analysis with a trusted uninvolved colleague. An additional review of the case and discussing recommendations may be a way to mitigate and manage the COI and help ensure that an EC’s conclusions are not unduly influenced by COI.

Fourth, the EC may recuse oneself from the consult and refer to another colleague. This is often the right choice when there are ongoing active dual responsibilities. This may also be the right choice with past or future dual roles for consults involving contentious disputes or that require mediation.

Fifth, the EC may refer the case to the IEC. Nationwide, a variety of team compositions for ethics consults exist. However, nearly all hospitals have standing IECs that by nature have greater numbers and more diverse training of members and there- fore should be less susceptible to individual biases or conflicts. Some IECs routinely ask members to support opposing sides of a complex case to help ensure all viewpoints are represented. Referral to the IEC may be right choice for more complicated or contentious consults as well as those that address recurrent concerns.

Sixth, the EC may refer to outside ethics expertise. This option is rarely selected, but should be available for intractable conflict when there is a complete loss of trust by a patient, family or clinical team. In these situations, referral to an EC or committee outside the reaches of the hospital may help foster a dialogue that would otherwise be impossible.

Our proposed algorithm We present our proposed algorithm in tables 3 and 4 based on our own experience and ethical judgement as ECs. In table 3 we specify points that accrue based on details of the ethics consult arranged by our four categories of items relevant to COIs for ECs. The score is then used in table 4 to determine the recom- mended action in response to the COIs.

5 of 7Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725

Original research

Importantly, our proposed algorithm is intended to be a starting point to guide ECs. Undoubtedly with use, areas of refinement will be identified. We have included ‘major’ and ‘minor’ classifi- cations in categories to give a degree of flexibility in application. Ethics consult services may feel that some risks are less appli- cable, while others are more concerning and therefore change points accrued for certain categories or alter the thresholds for proposed actions. Nonetheless, this is a starting point for consid- eration for a problem that currently lacks proposed solutions.

Application of our proposed algorithm We return to our original consult: the initial EC is asked to comment on permissibility of therapies (dialysis and renal transplantation) she is also responsible for providing. The consultant and requestor both recognised that while the initial consult simply reflected a question regarding permissibility of renal replacement treatment, there was growing clinician moral

distress as well as conflict between parents and the medical team. The competing roles of the nephrologist and the nature of the ethics consult create a COI and therefore action should be taken by the EC. While she has not yet met the family, she does have a clinical responsibility to care for the child as a nephrologist overnight and, if changes to dialysis were made or other treat- ments were initiated, she would be expected to have a close rela- tionship with the family until the child’s death or graduation from paediatric care. Recognising both the active current role and high likelihood of an in- depth future role, the EC chose to recuse herself and request a colleague to serve as the primary consultant for the consult. The nephrologist did provide tech- nical expertise as background and context to the next EC.

The neonatologist with a past dual role recognised the signif- icance of his previous interactions with the family and chose to recuse himself from the current ethics consult; this did result in losing the benefit of having an EC with increased specialised medical knowledge. As there was not a third EC available, the case was referred to the IEC for consideration.

Following our algorithm, the first EC received 11 total points for this consult (see shaded rows, table 5). The second EC had a different, but still relevant dual role and so received 8 points (see arrows, table 5).

Strengths and limitations of our proposed algorithm We suggest that ECs should be scrupulous in taking action to address COI and hope that the proposed algorithm can make ECs more confident in their decision about whether to take action based on COIs. The strengths of our proposed algorithm include the ease of application, practicality and allowance for flexibility for particular cases. Adoption of our proposed algorithm could increase transparency in the ethics consult process and thereby increase trust and buy- in from diverse stakeholders including patients, families, clinicians, the community and medical institu- tions. We hope such increased transparency and trust could lead

Table 3 Proposed scoring algorithm for conflicts of interest in ethics consults

Category Key features Points Example

Consult type Providing moral framework or providing objective answer

0 Team asks for guidance implementing Harm Principle in a particular case

Moral distress 1 Team is not asking for guidance on what to do, but wants support for bedside nurses reporting moral distress

Dispute 2 4 (intractable)

Team asks for guidance in settling dispute with upset surrogates

Dual role None 0 No concern with EC dual roles

Past 1 (minor) 4 (major)

EC was previously patient’s primary nurse

Active 5 EC is currently patient’s primary social worker

Future 2 EC will become patient’s primary attending next week

Team composition Committee 0 More than 4 ECs consider case before recommendations

Small team 1 Small team of ECs (2–4 individuals) serve together to make recommendations

Individual 2 Single EC considers case and makes recommendations

Other Interpersonal 1 (minor) 3 (major)

Early career neonatologist/ethicist is EC for consult requested by NICU division chief

Idiosyncratic position 1 (minor) 3 (major)

EC has published articles promoting idiosyncratic positions related to case

Bias 1 (minor) 3 (major)

EC has a fixed conclusion based on genetic diagnosis before meeting with stakeholders

Financial 1 (minor) 3 (major)

EC asked whether to order a very expensive test on which he holds a patent

EC, ethics consultant; NICU, neonatal intensive care unit.

Table 4 Proposed response algorithm for conflicts of interest (COIs) in ethics consults

COI score

Degree of concern about COI Response Continue EC role?

0–4 Lowest Take no action Yes 4–6

Disclosure only Yes

6–8 Obtain a second opinion

Yes

>8 Refer case to a colleague

No

>10 Refer case to the committee

No

>12 Highest Seek outside consult No

EC, ethics consultant.

6 of 7 Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725

Original research

to increased confidence in the ethics consult process itself. This is an empirical question which warrants further study.

Increased recognition and disclosure of COI does have possible drawbacks. The potential for loss of content expertise is inherent to a proposal that encourages taking action for COIs related to dual clinical roles. It is possible that such practices would discourage clinicians who occupy dual roles serving as EC. Similarly, we might lose insights from an EC whose norma- tive scholarship focuses on the topic of the consult.

Sometimes, recusal may be difficult or even impossible. In smaller institutions, there may be only a single available EC and therefore recusal may result in failure to provide an ethics consult.16 In such situations, the EC should make their best faith effort recognising limits of particular situations and create a process to address COIs when they arise. This may mean performing the consult but getting a second opinion, or pursing other potential options such as outside referral.

Although consult by larger groups or full committees should decrease risk of COI, it does not eliminate it and in some cases may be unhelpful. For example, if an individual in a leadership role has a COI and is part of a consult team, other members of the team may not feel adequately positioned to share relevant concerns.

COIs may also arise if the EC is perceived as beholden to the institution. It is true that all ECs will have some level of affil- iation, professional or financial, with the institution in which

they work. This relationship may become problematic if the EC feels pressured to inappropriately prioritise institutional needs or is perceived to do so by patients, families or clinicians. The institution as a stakeholder voicing concerns about liability or as the employer of the EC may also contribute to a COI. The more integrated an EC is into an organisation and its culture, the greater risk that COIs can result in errors of judgement.7

We do not consider cases in which an ethics consult is used inappropriately, such as in place of or as an extension of risk management. Hospitals have financial and reputational incen- tives to settle disputes.12 There may be perceived pressure towards supporting the institution or quickly resolving disputes even though to do so may not be the most ethically sound outcome. These issues are not within the scope of ethics consults so ECs should not be doing them.

We do not consider ‘organisational ethics’ or ‘institutional ethics’ consults in which there is not a particular patient or family of interest. These may have COI but are outside the scope of this proposed framework.

CASE #2: ADULT WITH AUTOIMMUNE ENCEPHALITIS An adult with autoimmune encephalitis remained in the inten- sive care unit requiring mechanical ventilation with aggressive management. Clinicians struggled to control the paroxysmal sympathetic hyperactivity ‘storming’ seizures with medications and resorted to the use of bite block and casting to limit injuries from the severe movement disorder the encephalitis caused.

Table 5 Case #1: neonate with neurological impairment and kidney disease proposed algorithm for conflicts of interest in ethics consults

Category Key features Points

Consult type Providing moral framework or providing objective answer

0

Moral distress 1

Dispute 2

4 (intractable)

Dual role None 0

Past 1 (minor)

4 (major)

Active 5

Future 2

Team composition Committee 0

Small team 1

Individual 2

Other Interpersonal 1 (minor) 3 (major)

Idiosyncratic position

1 (minor) 3 (major)

Bias 1 (minor) 3 (major)

Interpersonal 1 (minor) 3 (major)

EC #1 received 11 total points for this consult (see shaded rows) and therefore decided to refer the case to a colleague. EC #2 received 8 total points for this consult (see arrows) and therefore decided to refer to IEC as a third EC was not available. EC, ethics consultant; IEC, institutional ethics committee.

Table 6 Case #2: adult with autoimmune encephalitis proposed algorithm for conflicts of interest (COIs) in ethics consults

Category Key features Points

Consult type Providing moral framework or providing objective answer

0

Moral distress 1

Dispute 2

4 (intractable)

Dual role None 0

Past 1 (minor)

4 (major)

Active 5

Future 2

Team composition

Committee 0

Small team 1

Individual 2

Other Interpersonal 1 (minor) 3 (major)

Idiosyncratic position 1 (minor) 3 (major)

Bias 1 (minor)

3 (major)

Financial 1 (minor) 3 (major)

EC #1 received 13 total points for this consult (see shaded rows) and therefore decided to refer the case to a colleague. EC #2 had no relevant dual role and no relevant bias, so received 4 total points for this consult (see arrows) and therefore determined it was acceptable to perform the consult and no action was required related to COI. EC, ethics consultant.

7 of 7Weiss EM, et al. J Med Ethics 2021;47:e61. doi:10.1136/medethics-2020-106725

Original research

After months and little improvement, the clinical team requested an ethics consult to navigate the disagreements between the clinicians and family about the direction and goals of treatment. The patient’s attending and spouse wanted to continue aggressive treatment and the patient’s bedside nurses and parents wanted to transition to comfort measures. The EC on call had worked as a primary nurse for the patient multiple times since admission, had developed a relationship with the patient’s spouse and parents, and personally felt the patient was suffering each day of treatment and would be better served pursuing comfort care.

The EC recognised the COIs due to the past and potential future relationship with the patient and family, and the bias based on EC’s involvement with previously expressed support of comfort measures. Following our algorithm, the EC received 12 total points for this consult (see shaded rows, table 6) and therefore decided to refer the case to a colleague. The case did not change, but the second EC had no relevant dual role and no relevant bias, so the EC received 4 total points for this consult (see arrows, table 6) and therefore determined it was acceptable to perform the consult and no action was required related to COI.

CONCLUSION Here we have presented guidelines for ECs faced with COIs so they may make the determination of what, if any, action should be taken to lessen risks inherent to COIs. We suggest that ECs should be scrupulous in taking action to address COIs and hope that such guidance can make ECs more confident in their deci- sion whether to take action based on COIs. Our proposal is a first step: future work should attempt to validate the scoring system with bioethicists at multiple institutions, assess whether its use is of benefit to ECs, and refine it based on needs and prac- tices at particular institutions. Greater transparency in response of ethics consult services to COI is needed to increase the legiti- macy of this important work.

Twitter Laura Webster @laurabwebster

Contributors This project was initially conceptualised by EMW, AW and DD. LW helped develop the framework. EMW drafted the initial manuscript. All authors verify the final version as submitted.

Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not- for- profit sectors.

Competing interests None declared.

Patient consent for publication Not required.

Provenance and peer review Not commissioned; externally peer reviewed.

Data availability statement There are no data in this work.

ORCID iD Elliott Mark Weiss http:// orcid. org/ 0000- 0003- 2473- 9638

REFERENCES 1 Rosner F. Hospital medical ethics committees: a review of their development. JAMA

1985;253(18):2693–7. 2 American Academy of Pediatrics. Committee on Bioethics. Institutional ethics

committees. Committee on bioethics. Pediatrics 2001;107(1):205–9. 3 Guidelines for ethics committees in health care institutions. Judicial Council. JAMA

1985;253(18):2698–9. 4 Tarzian AJ, Asbh Core Competencies Update Task Force 1. Health care ethics

consultation: an update on core competencies and emerging standards from the American Society for bioethics and humanities’ core competencies update Task force. Am J Bioeth 2013;13(2):3–13.

5 Gasparetto A, Jox RJ, Picozzi M. "The Notion of Neutrality in Clinical Ethics Consultation". Philos Ethics Humanit Med 2018;13(1).

6 Parker JC. Clinical ethics consultation after God: implications for advocacy and neutrality. HEC Forum 2018;30(2):103–15.

7 Meyers C. Clinical ethics consulting and conflict of interest: structurally intertwined. Hastings Cent Rep 2007;37(2):32–40.

8 Fox E, Myers S, Pearlman RA. Ethics consultation in United States hospitals: a national survey. Am J Bioeth 2007;7(2):13–25.

9 Levine- Ariff J. Institutional ethics committees: a survey of children’s hospitals. Issues Compr Pediatr Nurs 1989;12(6):447–61.

10 Kesselheim JC, Johnson J, Joffe S. Ethics consultation in children’s hospitals: results from a survey of pediatric clinical Ethicists. Pediatrics 2010;125(4):742–6.

11 Frolic A, Chidwick P. A pilot qualitative study of "conflicts of interests and/or conflicting interests" among Canadian bioethicists. Part 1: Five cases, experiences and lessons learned. HEC Forum 2010;22(1):5–17.

12 Spielman B. Has faith in health care ethics consultants gone too far? risks of an unregulated practice and a model act to contain them. Marquette Law Rev 2001;85(1):161–221.

13 Frolic A, Chidwick P. A pilot qualitative study of "conflicts of interests and/or conflicting interests" among Canadian bioethicists. Part 2: Defining and managing conflicts. HEC Forum 2010;22(1):19–29.

14 Lyren A, Ford PJ. Special considerations for clinical ethics consultation in pediatrics: pediatric care provider as advocate. Clin Pediatr 2007;46(9):771–6.

15 Ginsburg S, Levinson W. Is there a conflict of interest? JAMA 2017;317(17):1796–7. 16 Tarzian AJ, Wocial LD, ASBH Clinical Ethics Consultation Affairs Committee. A code of

ethics for health care ethics consultants: journey to the present and implications for the field. Am J Bioeth 2015;15(5):38–51.

17 American Bar Association. Model rules of professional conduct, 2013. Available: https://www. americanbar. org/ groups/ professional_ responsibility/ publications/ model_ rules_ of_ professional_ conduct/ rule_ 1_ 7_ conflict_ of_ interest_ current_ clients/ comment_ on_ rule_ 1_ 7/ [Accessed 8 Jul 2020].

18 Sontag DN. Are clinical ethics consultants in danger? an analysis of the potential legal liability of individual clinical Ethicists. Univ PA Law Rev 2002;151(2):667.

19 Purtilo RB. Ethics consultations in the hospital. N Engl J Med 1984;311(15):983–6. 20 Frader JE. Political and interpersonal aspects of ethics consultation. Theor Med

1992;13(1):31–44. 21 FitzGerald C, Hurst S. Implicit bias in healthcare professionals: a systematic review.

BMC Med Ethics 2017;18(1):19. 22 Jones CP. Levels of racism: a theoretic framework and a gardener’s tale. Am J Public

Health 2000;90(8):1212–5.

© 2021 Author(s) (or their employer(s)) 2021. No commercial re-use. See rights and permissions. Published by BMJ.