Working through an ethical situation
A medical communicator who has recognised that something is wrong still has to decide what to do about it, and a code of conduct does not answer that. This topic carries a documented five-step process for taking a situation from recognition through to review, the teaching case it was built around, and the tactics writers use when the concern has to be raised with a client. What the standards themselves require is in scientific integrity in writing, good clinical practice as a standard and confidentiality and responsible publication; ending an unworkable client relationship is in when a client relationship fails, and the workshops the model was written for are in association workshops and certificates.
What ethics means
Ethics means two things. One is principles of conduct that distinguish between right and wrong. The other is the scholarly study of right and wrong, which addresses the institutions that have established a society's standards and rules for how people treat their fellow human beings and, more recently, the other inhabitants of the planet and the environment, and for making sure that what people do has as few unintended negative consequences as possible. Acting responsibly means thinking as a chess master does, several moves ahead, so that the potential downsides of a decision are anticipated and do not come as a surprise.
Expect the situations to arrive from any part of the work
The process and its teaching case come from an article in the first issue of the AMWA Journal's 2012 volume by Cindy Hamilton, principal of the consultancy Hamilton House in Virginia Beach and one of the American Medical Writers Association workshop leaders who developed the model. The four AMWA Journal issues this topic draws on, numbers 1 and 4 of the 2012 volume and the Fall and Winter 2017 issues, were each converted from PDF by layout-aware Poppler XML conversion and admitted without content repair, so figure labels can be inlined into prose, cover and contents columns can interleave, and reference URLs can be malformed or truncated.
Ethical situations are not a publication problem that other specialisms escape. They keep changing because medical communication keeps changing — in subject matter, in the services offered, in the media used and in the settings people work in — and they can involve science or medicine, research or clinical practice, or another subject entirely; writing, editing or developing materials in any medium; and work inside a pharmaceutical company, a medical communication company, a health care facility, an independent freelance business or somewhere else again. Some of them are obvious while how to handle or avoid them is not, which is the gap a process is for.
Others arrive as a feeling about a manuscript rather than as a question anyone has asked. Kelly Keating's summary of a session on retractions, printed in the fourth issue of the same 2012 volume, opens on four such thoughts: something does not seem quite right here; this seems familiar; why does the writing suddenly improve in this paragraph; is the image in this figure distorted. A writer or editor who has had those thoughts and then followed them up by diplomatically querying the author, the summary runs, can add ethical detective to their list of skills. What that session advised doing next is set out with the tactics for raising a concern below.
In writing for patients and the public, one of the biggest ethical concerns is the source of funding and information. Medical writers must convey scientific information to patients and the public accurately, fully and non-promotionally while still meeting the sponsor's requests and needs for a project, and the needs of the public and of a sponsor can seem at complete odds with one another. Writing is a large part of the job, but learning to compromise successfully on complex topics to provide useful documents is more the task at hand, and a medical writer must take seriously the duty of being one of the checks and balances on all scientific information that reaches the public.
Work the case rather than read it
The model is taught against a single case report assembled from the real experiences of two association members, combined and used with their permission, who preferred not to be named. Vera City, an experienced writer and long-standing member newly employed full-time by a start-up medical communication company, was assigned an article about cancer care at a large health care facility for a local magazine for seniors. The client supplied marketing and public relations material about the facility's novel design, its technologies and its model of care which, taken together, made the facility seem extraordinary and implied better outcomes for its patients than other facilities achieved.
Her own research found several solid sources indicating the claims were not substantiated, and, as she dug further, that the outcomes the client's materials advertised were no better than those elsewhere. She put the findings first to her supervisor and then to her contact at the facility and was ignored both times; uncomfortable but continuing, she wrote an article she hoped reflected the facts in a tone the client would accept, and her supervisor rewrote it to satisfy the facility's marketing team. She stayed at the company and moved on to other projects, but kept returning to it: driving past the facility she pictured sick but hopeful patients who might have chosen a treatment on the strength of misleading marketing material. Should she have done anything differently, and should she act now?
Hamilton's instruction is that reading the case is not the intended use. The reader is asked to imagine sitting in a workshop with members from different specialties, work settings and levels of experience, and to simulate that by discussing the case with colleagues and customising it for their own specialty and setting, working each step of the model through the discussion. The reason given is that, because the nature of the work keeps evolving, everyone needs advice and reminders about best practice.
Recognise the situation, and define it before analysing it
The first step is to establish that an ethical situation exists at all, and the case is chosen so that the answer is not obvious. Vera had already acted: she raised her concerns with two parties and wrote an article that reflected the facts. Whether those actions were sufficient is exactly what is in question, and the article sets against her having moved on to other projects the fact that she continues to think about the experience.
Defining the situation as clearly and succinctly as possible is the harder half of the step, because real situations are complex and admit many perspectives. The definition the article settles on for the case is one sentence about the outcome rather than about the conduct: ethical behaviour did not result in a final product that was objective, scientifically accurate and rigorous, and fairly balanced.
Then identify everyone involved and everyone who may be damaged or otherwise affected, directly or indirectly, which extends the list well past the parties to the argument. Patients with cancer might have made treatment decisions on the strength of misleading marketing material. Patients without cancer could be affected if they generalise from the article, and more so if they later learn it was misleading — to the point of questioning the benefit of the facility's non-cancer treatments or, worse, of any medical treatment at all. Reputations are also at stake: the medical communication company's, the facility's, and those of employees of both who had nothing to do with either the article or the delivery of cancer care.
The step closes on why the decision matters to the person facing it. In the case, she wants to learn from the experience, which is why she is asking both whether she should have acted differently and whether she should act in future.
Investigate the facts, your intention, and your own conflicts
The second step separates three things that are easily run together. The facts come first: distinguish them from opinions, desires and other wishes, name the assumptions that have been made, and ask how certain you actually are — in the case, how certain Vera is that the article is misleading, and whether her own research could have been incomplete or otherwise flawed.
Intention is a separate question and is asked of yourself. Is the motive what is best for you, for your employer, for the profession, for the patients nearby, for society as a whole, for some combination of those, or for something else again?
The conflict-of-interest question is where the step opens furthest, because most communicators are familiar with the financial conflict that an author has to disclose. The article's list of what else counts is worth carrying into any situation: a personal conflict arising from your own experience — chemotherapy that caused irreversible toxicity, or a medication error at the organisation concerned that killed a family member; a romantic conflict from a current or past relationship with someone at either organisation; and religious, cultural, ethnic, political, loyalty and other conflicts besides.
Only then does the step reach the external sources: the relevant statutes, regulations and guidelines, and the association's code. The case engages the first three principles of the AMWA Code of Ethics, and the third is the pointed one, because it requires information meeting the highest professional standards whether or not the material comes under any regulatory agency's purview — and Vera's article, Hamilton notes, was probably not subject to regulatory review.
The step then asks how to find out what is still unknown, and what further guidance professional industry guidelines, peers, colleagues and ethicists can give. Does the organisation have guidelines about marketing materials, or an ethics hotline? Does it belong to a body that has such guidelines? Can fellow association members advise? And the step ends on causes rather than blame — why is this happening at all? The case's candidate answers are that a start-up has not yet had the chance to write standard operating procedures for reviewing deliverables, and that the facility may be unaware of its own marketing guidelines.
Gauge the options before judging any of them
The third step begins by settling exactly what must be decided, and it redirects the question forward. Most of the questions a person in this position asks are about what they might have done differently; considering alternatives to past actions has educational value, but focusing on future actions may be the more productive use of the analysis.
List every possible course of action, at length, and resist making judgments about them while the list is being made. Most lists begin with doing nothing, which feels counterintuitive and is almost always a possible action. The case's list asks how Vera can manage the risk to herself and others, and runs: perform additional research to confirm the assumption that the article is misleading; document the attempts to raise concerns and keep her own version of the article; offer to help the medical communication company develop a standard operating procedure that includes a review process; offer to help the facility develop a policy for preparing marketing materials; and ask colleagues to identify further courses of action.
Deciding then means weighing the likely consequences of each alternative including its effect on each stakeholder, recognising that a single action is rarely ideal for all of them, and identifying the course — which may require several actions — that is best for the majority. Three tests close the step: does the decision maximise benefit and minimise cost and risk; have you treated others as you would want to be treated; and would you be comfortable if your reasoning and your decision were made public?
Implement the decision, then go back to it
The fourth step is about execution and asks when the decision should be implemented and who should act on it. Where the situation has already happened, as in the case, immediate action is probably not required and progressive corrective actions can be implemented gradually over time. Whether the person deciding is the right person to implement it is a separate question, and if not, someone has to be found who can or should assist — a supervisor, a client contact, or another party. The step finishes by checking that the proposed course has been communicated to all the interested and affected stakeholders, and that it is being implemented in a way that maximises benefit and minimises cost both to them and to the situation.
The fifth step comes after the decision has been implemented, and it is to evaluate it: was the situation handled successfully, how well was the decision implemented, what effect did it have, should the behaviour or the decision be modified in the light of that effect, and what can be learned for future situations?
Carry the five steps and six questions as a checklist
Put end to end, the five are: recognise the ethical situation, define it clearly and succinctly and identify all stakeholders; investigate the facts and assumptions, especially the relevant statutes, regulations and guidelines and the association's code, and identify every conflict of interest including the personal ones; gauge the situation and decide, by listing all possible courses of action and choosing the one or several that maximise benefit and minimise cost and risk for the majority of stakeholders; handle the situation and implement the decision; and tailor the decision by evaluating it for lessons learned and revising it as needed. The retraction session offered medical communicators the same five as a checklist for handling a difficult situation.
Andrea Gwosdow, who moderated that session, suggested six questions to ask yourself beside the checklist: who is involved in the situation and who will be affected by the decision; which principles in the association's code of ethics are relevant; what the conflicts of interest are; what all the possible courses of action are and the likely consequences of each; who can or should help; and what lessons can be learned. Using the checklist and the questions, the summary concludes, may help a writer educate clients and help them avoid unintentional ethical lapses or, conversely, detect intentional wrongdoing and decline to work with the client any longer. The conversion sets the end of that sentence, and the summary's closing lines, inside the neighbouring summary of a session on writing books, so the join between the two halves is read across that break.
The model was still in use five years later. Tamara Ball, a principal medical writer and the editor of the AMWA Journal's Everyday Ethics section, summarised it in her column in the Fall 2017 issue, and her wording adds detail to the steps. Recognising includes naming the issue, identifying who is directly involved and listing every stakeholder who could possibly be affected by the final decision. Investigating includes identifying the relevant ethical codes, statutes, regulations and laws, taking stock of your own competencies and biases, and considering potential conflicts of interest. Gauging is the step in which decisions are made, by listing every potential action and its likely effect on each stakeholder, the best decisions maximising benefit and minimising cost and risk. Handling means that, once the what and the why have been determined, the communicator considers who, when, where and how the decision will be implemented. And tailoring is a backward look over the experience for lessons to carry forward.
Borrow the tactics writers use with a client's data and authors
Two records of practitioners' advice describe how to raise a concern with the people whose manuscript it is, and they agree on the posture. At the retraction session Gwosdow's advice was to work in a collaborative spirit with clients, educating and informing, so as to head off any tendency toward plagiarism or falsification. Misconduct can sometimes be unintentional — her illustration is a graduate student contributing data who may not know that it is not permissible to erase small bits of debris or noise from an image — so an innocent-until-proven-guilty approach is described as the most diplomatic, and a writer corresponding with an author should tread carefully and respectfully, asking questions without accusing, to get to the bottom of any suspected misconduct.
The fuller playbook was recorded from the second Meet and Share session of EMWA's medical communications special interest group, held in November 2021, by Sampoorna Rappaz, a freelance medical writer and editor based in Switzerland, in the March 2022 issue of Medical Writing. The situations are familiar: on reading the data with fresh eyes, the writer finds exaggerated conclusions, problematic analysis or poor data quality, authors who have not contributed enough for a byline, or illegitimate authorship requests at submission. The March 2022 and June 2022 issues of Medical Writing (volume 31, numbers 1 and 2) drawn on in this topic were admitted from an existing Markdown conversion without content repair; their known limitations are broken ligatures, mangled multi-column tables, interleaved page headers and footers, and figure text represented as marked text.
The recommendations are to communicate risk — tell a sponsor whose conclusions are inflated what it stands to lose if its audience's trust goes, since a loss of reputation would affect brand and market value, while acknowledging the good faith in which research is conducted, a discussion the record calls harder with clients who feel personally involved in the study; to communicate politely and clearly, asking questions and requesting clarifications in a helpful, respectful and humble tone and offering actionable suggestions such as a root cause analysis when a primary outcome analysis disappoints; and not to let impostor syndrome silence doubts, advice the record calls especially pertinent to junior writers and which the experienced writers said most publication teams welcome when it is raised inoffensively — adding that a client's reaction tells the writer whether to work with them again, and that losing a few business prospects in the short term is better than attracting the wrong clients. Do your own calculations: check that the limits of data ranges are sensible, units accurate and basic calculations correct, and look at minimum and maximum values for outliers; basic statistical knowledge helps, and a meticulous nature finds errors too. Make it less personal by pointing to best-practice guidelines and to the regulators, reviewers and editors who will raise the same concerns, with questions such as what the client would not want a regulator to ask, whether anything in the data could embarrass them, whether the dataset is verified, and what the editor or peer reviewer will say.
Share responsibility: with editors and reviewers, by stating limitations and any concerns affecting the conclusions; with statisticians, who should own data quality and analyses and merit authorship — editorial boards and reviewers are sceptical of analyses with no statistician author, which is an argument for the client; and with authors, by taking issues to the lead author. On authorship, the writer is the first line of control that listed authors meet the criteria: make silent authors take ownership by asking questions that need detailed answers, share inclusion and exclusion criteria (ICMJE and GPP3) with authors and sponsors early, align with departments affected when a prominent author is removed, and do not succumb to commercial pressure. A process-driven eligibility form based on CRediT contributor roles, agreed at kick-off, updated throughout and finalised at submission, can justify authorship to an editor, settle disputes and serve as evidence at audit, alongside the records of publication-planning software such as Datavision. How the authorship criteria themselves are applied is in deciding who qualifies as an author.
Both records say where the effort can end. If the issues cannot be resolved, the EMWA group's last recommendation is to walk away and ask for your name to be removed from the acknowledgements. Rebecca Lew, a senior medical writer who also spoke at the retraction session, suggested that medical writers can act as the final gatekeeper and by their diligence help protect themselves, their clients and the journals, and contribute to accuracy in the medical literature. Her advice at the same session never to cite a retracted article is in checking whether a paper you cite has been retracted. The EMWA group's own summary is that confidence, clear positive communication and shared accountability are what reach the ethical objective.
Know where the model came from
The five steps are known by the mnemonic RIGHT, which Peggy Boe chose for them, and the model was developed by the leaders of the association's ethics workshops with the explicit goal of something simple, memorable and relevant to all medical communicators rather than to one specialism. The figure setting it out is credited to Marijke Adams, Tami Ball, Peggy Boe, Charmaine Cummings, Andrea Gwosdow, Cindy Hamilton, Jill Shuman and Nancy Taylor.
It was not invented from nothing. Hamilton records that other models were researched before five steps were settled on, and her reference list names three: the American Psychological Association's Ethics Desk Reference for Psychologists of 2008, the Markkula Center for Applied Ethics' 2009 framework for thinking ethically, and an ethical decision-making model published by the Josephson Institute of Ethics in 1999.
Hamilton discloses that she wrote the self-study module named in the article's figure of ethics workshops and receives royalties on it.
Ball's Fall 2017 column proposed to explore colleagues' experience through the model, which she describes as developed by the AMWA workshop leaders who taught ethics and experienced by many members. Her caution is that ethical thinking should be guided, but not shackled, by the code, whose proper use means careful consideration of the unique situation in question. She invited members, newcomers and seasoned alike, to send thoroughly anonymised cases — the essential details, the stakeholders, proposed solutions and, where the situation had been handled, how and with what result — preferably worked through the model, as a springboard for discussion in the column.
The column's second instalment, in the Winter 2017 issue — bylined Tami Ball, who refers back to her inaugural column — repeated the invitation with the same checklist: the essential details, the stakeholders, proposed solutions and, where the situation had been handled, how it was handled and what followed, with cases asked to follow the RIGHT model. She describes the interactive column as a new way of exercising the association's ethical muscle, one that might renew awareness of the principles governing the profession, and presents members' willingness to share their knowledge and experience for each other's benefit as a benefit of membership.