Communicating risk and uncertainty
A reader deciding whether to accept a treatment, a vaccine or a screening test is not weighing a probability in the abstract; they are weighing a small chance of a harm they can picture against a larger chance of a harm they cannot. Writing about risk is therefore a distinct skill from writing about numbers, and this leaf answers how to write a risk so that such a reader can weigh it: what to set it beside, what the reader already brings to it, how the wording, the depth and the format change what is understood, and how trust is kept. The arithmetic underneath — absolute and relative risk, and how a treatment effect is expressed — is in risk and treatment effects; the reader hesitant about a vaccine in particular is in writing for a vaccine-hesitant reader; and how much a reader can do with health information at all is in health literacy.
Put the risk beside the risk it is being weighed against
Several of the examples in this leaf come from a 2007 AMWA conference session on vaccine safety in which a science writer, an immunisation-programme director and a vaccine-policy physician described what does and does not work. Vaccines supply those examples; the technique generalises to any communication where a rare harm is set against a common one.
The session's central instruction was to use risk communication rather than plain explanation: if the writer compares the risk of an adverse event against the risk of the disease, readers make better decisions. A bare adverse-event rate, however accurately stated, invites the reader to compare it against zero, and zero is not the alternative on offer.
Part of why the comparison is needed is that the alternative has become invisible. Where a disease has been suppressed by successful vaccination, people with no memory of what it did to earlier generations do not fear it, so the only vivid quantity left in the decision is the intervention's own risk. The session closed on that point: many people are not worrying about the diseases, and the writer's job is to communicate real risk rather than let misinformation fill the gap.
A radiation dose is the standing case, because the number means nothing on its own to anyone who does not work with it. Answering the perception that dentistry overuses radiology — a stigma she reports among writers, journalists and the public — the dental-radiology speaker at EMWA's May 2024 expert seminar on AI reaches for food rather than for a unit: three Brazil nuts, which carry a thousand times more radioactive radium than other foods, deliver the equivalent of one intraoral radiograph of the kind taken in private practice every day. The figures are hers and the record is uncorrected automatic captions, so read them as the shape of the comparison rather than as measurements. What the technique does is convert an unfamiliar quantity into a familiar one the audience can already weigh, and it works because the comparator is genuinely everyday rather than chosen to flatter.
The same comparison can be used to make a familiar risk visible again rather than to shrink an unfamiliar one. An injury prevention centre director told a 2009 conference session that injury prevention is popularly associated with disastrous events such as terrorist attacks and tornadoes, while the data show more deaths from every manner of preventable injury in a given place and period than from the catastrophic event that filled the newspapers for weeks; her centre's work was to reframe the perception so that traffic deaths are weighed with the same gravity as deaths in a war then being reported daily, and to act on specifics — raising car seat and seat belt use in neighbourhoods with high rates of non-use, and the number of smoke alarms in areas with frequent residential fires. The technique is the vaccine one run the other way: the comparator is chosen because the audience already takes it seriously.
Know what the reader brings to a risk
Fright is not proportional to probability, and the features that drive it are predictable enough to plan for. The vaccine-safety session's list of what alarms the public is risks that are involuntary or inescapable, risks that are man-made, risks potentially damaging to children, and situations in which contradictory statements are issued. Three of those four describe a routine childhood immunisation exactly, which is why vaccine communication is hard in a way that communication about a comparable voluntary risk is not. The fourth is the one the writer controls: contradiction between sources is itself a fright factor, so inconsistency across a campaign's materials does damage independently of what any single document says.
Fright is one thing the reader brings; faulty assumptions are another. A 2009 AMWA conference session on risk management in the pharmaceutical world approached them from the regulatory side, and its framing is worth having because it puts communication inside a cycle rather than at the end of one. Jennifer Grodberg, a regulatory affairs director who moderated the session, described risk management as iterative: assess the benefit-risk balance, develop and implement tools to minimise the risks, evaluate whether the tools worked, modify them, and return to the assessment. What the session then examined was the part of that cycle carried by communication — how public risk is reduced through timely, adequate and informative communication by and to consumers, physicians, pharmacists and providers.
Jane Neff Rollins, a medical writer at a communications agency, located the fundamental challenge before any of that: the public's perceptions of risk in health care are faulty, and at the root of it is a rampant innumeracy that she calls the mathematical equivalent of illiteracy. Children are not taught to interpret probability and statistics at school, public science education is grossly inadequate, even television coverage of science was declining, and consumers are taught nothing about drug research and development or about what the approval process does.
In the absence of that education people make assumptions that are natural and wrong, and the list is worth having because each item predicts a particular misreading. Minor happenings are ascribed to chance and major ones to a cause, so tripping is an accident while a serious adverse event must have been caused by the drug. Natural is equated with safe. Risk is believed to be avoidable altogether by behaving in certain ways. Pseudoscience is accepted on the same footing as rigorous inquiry. And the regulator's ability to keep people safe is distrusted, with the distrust increasing when a product is taken off the market for safety reasons — which is one of the agency's critical functions rather than a failure of it.
Underneath the list is a discomfort with ambiguity, uncertainty and the weighing of risks against benefits. Without a working picture of the scientific method, a reader does not expect old data to be displaced routinely by new studies: scientists and health professionals understand themselves to be in a continual evolution toward truth, while the public, in Neff Rollins's borrowing from Stephen Colbert, sees only truthiness. The documents the agency regulates make that worse rather than better, being pitched far above the people who have to read them — readability measures and reader testing carries the grade-level comparison — and her conclusion is that health communication has to catch up with a less paternalistic model of the patient-physician relationship, in which patients take a more direct role in their own care.
Where the risk being explained is a vaccine, the reader brings something more specific than innumeracy, and the research on it says that supplying the missing numbers is not what changes a mind. Individual decision-making about vaccination involves emotional, cultural, social, spiritual and political factors as much as cognitive ones, so the knowledge-deficit assumption behind most vaccination campaigns has repeatedly failed to move the people it was aimed at. Writing for a vaccine-hesitant reader collects what has worked instead.
A rhetorical analysis presented at AMWA's 2013 annual conference approaches the same problem from the clinician's side and says something usable about why supplying the numbers fails. Its subject is how doctors argue in controversial medical situations — from the debunked link between the MMR vaccine and autism to the annual questions about seasonal influenza vaccine — with paediatricians in particular having to persuade hesitant parents about a range of diseases carrying different consequences. The data were qualitative interviews with physicians together with professional publications from the American Academy of Pediatrics database, analysed by rhetorical methods: the rhetorical situation, and appeals to credibility, to emotion and to facts. What the analysis found is that forms of evidence, value systems and definitions of health shape how persuasive a standard medical recommendation is, and that doctors communicate about vaccines from particular notions of risk, benefit and family or community responsibility that may not correspond to, or account for, what individual parents are actually worried about. Its conclusion is that rhetorical analysis shows how, why and under what circumstances scientific appeals fail to persuade particular lay audiences — a diagnosis rather than a script, and one that names a mismatch of premises rather than a deficit of information. The abstracts in that issue were published exactly as submitted, without the journal's copyediting. The issue was admitted from an existing Markdown conversion without content repair, in which figure labels can be inlined into prose, cover and contents columns can interleave, and reference URLs can be malformed or truncated.
Answer the individual question the average does not
"What if my child is one in a million?" is described in the vaccine-safety session as a real concern rather than a misunderstanding of probability, and the response prescribed is to communicate with compassion and empathy toward people's beliefs. A population rate is a true answer to a different question from the one being asked, and a writer who supplies only the rate has not answered.
The same asymmetry has a collective form the writer should be able to state plainly. Population protection depends on a threshold proportion being immune — above 94% of the population for a disease as contagious as measles — and below it outbreaks reach people who cannot be protected individually, including the immunosuppressed, children too young to be immunised, and the roughly 5% of measles vaccinees in whom immunity does not take. One speaker's formulation of this was that the immunity of his neighbours determines the safety of his grandchild; another described parents who avoid the individual risk while depending on others to keep the population disease-free. Whether to put that argument to a reader who has just asked about their own child is a judgment about the document's purpose, but the writer should know it is available and what it costs in tone.
Word the risk so that the framing does not decide for the reader
Two pieces of EU legislation turned the wording of benefit and risk from a communication preference into a writing task, and Lisa Chamberlain James, a senior partner at Trilogy Writing and Consulting, uses them to set out what the underlying research says. Human decision processes are limited by heuristics and biases, the effects of which are greatest in decisions involving risk or ambiguity; the amount and type of bias differ between individuals and weigh most on people with low numeracy. How a treatment's risks or harms are framed — logically equivalent choices described in different ways — and the context in which they are explained can therefore be decisive in what a reader understands and in the outcome that follows. The standing example is survival against mortality: a treatment described as carrying a 90% chance of survival is judged more attractive than the same treatment described as carrying a 10% chance of death. The full impact of framing is difficult to assess, but the writer still has to decide how the framing is done and how much context to give to counteract the bias.
Individual words carry the same effect. An effect described using the word percentage is perceived as larger than the same effect described as reduced by or relatively reduced. Most people are risk averse to the point of choosing a less effective treatment they believe to be safer, and some avoid deciding at all where ambiguity is involved — and a term that is poorly understood, or not understood, becomes ambiguous. That is the argument against reaching for statistical or complicated medical terms in lay writing: using confidence intervals to convey how much trust a reader can place in a result can increase their perception of the risk and make them reluctant to take the medicine.
The 2009 risk-management session gave a list of things to do in the text, and its first item meets that ambiguity aversion head on: acknowledge ambiguity, and then say where the preponderance of evidence lies. Balance anecdotes with statistics. Use risk comparisons and risk-benefit analyses to put a risk in perspective. Use well-designed graphics to reinforce the concept visually. Show empathy for the public's perceptions of risk even where those perceptions are inaccurate, and correct misperceptions without patronising or ridiculing anyone. And let the public see the dilemmas faced and the reasons behind policy decisions rather than only the decisions themselves.
Melissa Christianson, a medical writer and consultant at Whitsell Innovations, gave AMWA's virtual 2020 conference the same research as obstacles and strategies, in a session reported by Núria Waddington Negrão in the Spring 2021 AMWA Journal. The Spring 2021 AMWA Journal issue (volume 36, number 1) was converted from its PDF by layout-aware Poppler XML conversion and admitted without content repair, in which figure labels can be inlined into prose, cover and contents columns can interleave, and reference URLs can be malformed or truncated. Negrão opens on the disconnect between perceived and actual risk — people fear terrorist attacks, gun crime and bird flu, while cardiovascular disease and cancer cause the most deaths — and Christianson defined risk as the probability of an event or outcome, and risk communication as the open, two-way exchange of information and opinion about risk leading to better understanding and better clinical decisions about management, with no clear right answer. The writer's job, in her account, is clarity in presenting incidence (new cases), prevalence (total cases), absolute risk (the probability of an outcome in a population) and relative risk (a comparison of the absolute risks of two populations).
Her four obstacles are that the maths is hard, many people, college-educated ones included, lacking the numeracy for the basic computations, which leads to overestimated risks and inconsistent readings of equivalent data; that people assess risk inconsistently, swayed by the statistical measure, the framing, mental shortcuts, emotions and their own numeracy and literacy — absolute risk is understood better than relative risk, while relative risk inflates the perceived effect and so treatment acceptance, and the report's framing example, citing McNeil and colleagues' 1982 study, is that more patients chose a treatment when told that 32% of patients were dead a year after it than when told that 68% were alive; that doctors and scientists speak a different language from the public, so that "significant" is normally read as clinically significant; and that personal beliefs enter the mix. Her six strategies: start basic, assuming no background and defining the risk, the time interval and the population; make the maths easy, with natural frequencies (2 out of 100 people), numeric and verbal descriptors together (low risk — 10%), absolute risk alone or beside relative risk (the risk of A was 10% and of B 20%), visual aids, and a build-up through levels of precision; be consistent between treatments in statistical measure, population, time interval, denominator and framing; give context and handle comparisons carefully, considering an incremental risk format that presents the background risk of an outcome regardless of intervention; reduce the emotion by removing emotive language, balancing the framing and being open about what is and is not known; and write to the audience's socioeconomic make-up, concerns, habits and knowledge, intervening proactively where they will have trouble assessing or acting on the risk.
Give context, and let the reader choose the depth
Risk–benefit information can be very difficult to explain, and the recommendation from a 2013 EMWA panel on patient information was to stop trying to find the one right depth: give different levels of information so that patients can decide for themselves how much detail they need. The same panel singled out numbers needed to treat as a figure that has to be communicated very carefully and explained so that it is put in context. Its closing observation on what the exercise is for is worth keeping in view — an informed patient is not always an obedient patient.
Two panel members gave the reasons behind the recommendation. Eveline Wesby-van Swaay, a global safety physician, put the difficulty down to the variety in patients' ability to understand and interpret medical data, and argued for tailoring information to individuals as far as possible. Ben Bridgewater, a cardiac surgeon whose own field publishes its surgical outcomes, reported that the public has a huge appetite for scientific and medical data and that the transparency in his area had been followed by decreased mortality rates and improved cardiac outcomes — with the qualification that information must be contextualised for patients and not handed to them as raw data.
The panel's worked example of the cost of getting this wrong was the fall in MMR vaccination rates. Wesby-van Swaay's reading was that if Andrew Wakefield's study had been properly explained to the general public the vaccine might not have met such a negative response; the panel's version was that none of the caveats or assumptions in that study were explained, so the public did not have the full story on which to make an informed decision. Read that as the panel's argument about the value of explanation rather than as a settled account of what drove the fall.
Chamberlain James makes Bridgewater's point about raw data from the side of the numbers. Giving a lay audience complex numerical benefit-risk information is often counter-productive and can impair rather than support decision making, which is why disclosure without context is not automatically a gain. Her hypothetical makes the point about baselines: publishing that 30% of participants in a trial reported their leg turning blue temporarily sounds alarming enough to deter someone from a medicine or a future trial, but if 29% of the general population have blue legs from time to time anyway, a 1% increase is neither frightening nor serious. Reading the figure correctly depends on knowing the background rate — which is exactly what a stand-alone results disclosure does not supply. Her conclusion is that empowering patients means more than publishing trial results or a summary of a risk management plan: given without context and in a form the reader can interpret, transparency can do more harm than good.
A regulator's safety notice builds the same recommendation into a fixed format, with a layer for each audience that it can stop at; the worked example below sets it out.
Choose a format and test that readers understand it
How a risk is expressed is a research question in its own right, and worth knowing about because the answer is not settled. Jessica Ancker, interviewed in the AMWA Journal in 2007 while working on it in a biomedical informatics department, describes her subject as how to express probability and risk information in non-word formats — graphs and interactive educational computer programs — and the problem motivating it as the finding that people still misinterpret individual risks despite a great deal of research in the area. Her example of the difficulty is deliberately ordinary: when people are told they have a 10% risk of something happening, do they really understand what that means?
The formats already tried divide into two families, and she was working on a third. Some earlier work described risks in words and some used graphics such as bar charts and stick figures; her own approach was interactive, developing a computer game to help patients understand their risk of a particular disease or condition, prototyped and then tested in focus groups. What sits underneath that choice is cognitive science rather than design preference — the course that drew her into the field was about how cognitive science explains aspects of human reasoning about complicated information. Two things transfer for a writer with no research budget. A verbal statement, a graphic and an interactive tool are three different interventions rather than three presentations of one thing; and whichever is chosen, whether the reader understood it is a testable question, which is why focus-group testing appears in the research and belongs in a serious patient-information project too.
The same research programme was expected to feed two other document types, which is worth registering because it says where the technique goes. Ancker names general health education materials and informed consent as the areas her work on risk comprehension would also help. Both ask a reader to weigh a probability and act on it, and both are governed by requirements that say what must be disclosed rather than what must be understood. Informed consent materials covers that gap from the regulatory side.
Where the same kind of decision recurs — how well does this drug work, and what does it do to me — a standard format can be designed once, tested with readers as Ancker was testing her game, and reused, and the drug facts box is the worked attempt at that. Lisa Schwartz and Steven Woloshin of Dartmouth Medical School set it out at the National Academy of Sciences' Arthur M. Sackler colloquium on the science of science communication, held in Washington on 21 and 22 May 2012, by first characterising the two channels that already existed. Commercial communication, such as direct-to-consumer drug advertising, is appealing, fun and easy to understand, and can mislead. Public communication, such as the Food and Drug Administration's approval documents, is more comprehensive and more useful, and is often poorly organised, weakly summarised, dense and frustrating to read.
Their third option borrows its design from the nutrition facts label on food packaging: a simple tabular layout displaying the data on a drug's benefits and its side effects. They tested it extensively to see how people reacted, and their results were that consumers both want the box and can understand it. The comparative finding is the one that matters for anyone arguing about format: against traditional drug advertising and against the laundry list of side effects, the box enabled people to choose the more effective of two similar drugs — including participants whose education had ended at high school or earlier.
The postscript is a lesson about the distance between evidence and practice. The drug facts box had by then been written into United States legislation, at section 3507 of the Patient Protection and Affordable Care Act, and implementation was still expected to take three to five years. Schwartz's comment on that gap is worth keeping beside any well-evidenced format proposal: getting more science into communication is one thing, and getting science communication implemented is another.
A standing reference exists for this whole subject and it is free, which is worth knowing before commissioning a literature search. The US Food and Drug Administration published Communicating Risks and Benefits: An Evidence-Based User's Guide and made it available online at no charge; its introductory chapters cover the goals of risk communication, methods for evaluating a communication, standards for judging whether one is adequate, and the language used, and later chapters take up how people interpret quantitative information and how emotions — and age — affect risk perception. Each chapter is self-contained, referenced and annotated, so it can be read as a single answer to a single question rather than end to end.
Read a regulator's own safety communication as a worked example
A drug regulator publishing an emerging safety risk faces every problem in this topic at once, and the US Food and Drug Administration's Drug Safety Communications are documented in enough detail to be read as a worked example. Laura Pincock of the agency's Safety and Risk Communication Team, speaking at a writers' conference in 2012, described the team's job as informing the public about new and emerging potential safety issues arising after a drug is marketed, within a mission to provide timely, understandable, relevant and actionable information so that health care professionals and their patients can make the best individual treatment decisions.
The distinction the whole genre rests on is between established and emerging risk. Product labelling — the package insert, legally required for every prescription medicine — is the agency's principal tool for communicating established risks, and adding a side effect or warning to it requires a high standard of evidence. Emerging risks carry greater uncertainty and may therefore not yet be in the label at all, which is what the separate communication exists to carry. Why such risks appear after approval is worth stating in the copy rather than assuming: approval rests on finite numbers of patients meeting strict eligibility criteria, and once a drug is outside the trial setting it may be taken by a different population or for a different disease, so the likelihood of an adverse event may actually increase.
What triggers a communication is a judgment about benefit and risk rather than a threshold on the risk alone. The agency's experts weigh the benefits patients get from the medication against the emerging risk, and four named factors move that balance: the seriousness of the adverse event relative to the benefits of the overall treatment, the strength of the evidence for a causal relationship, the potential for preventing or mitigating the risk, and the availability of alternative therapies — so where no other drug treats a serious disease, a higher risk may be accepted. The communication's primary purpose follows from that: to help an individual make an informed decision about their medication, not to stop them taking it, since patients do not necessarily need to discontinue a valuable therapy in the light of updated safety information. The July 2012 communication on a multiple sclerosis drug is the shape to copy: seizures were already a known side effect, and what the notice added was that the risk was greater in patients with reduced renal function, with the actionable instruction that prescribers check renal function before starting the drug.
The format answers the multiple-audience problem by tiering rather than by choosing. The communication opens with a Safety Announcement of a few paragraphs summarising the whole notice in consumer-friendly plain language — "the introductory paragraph is key and serves to hook in readers", in Pincock's words — beside a box giving facts about the drug such as its benefits and, where known, the number of patients who have used it. Additional Information for Patients follows in plain language, then Additional Information for Healthcare Professionals with more descriptive medical and technical content, and finally a Data Summary, the most detailed section, which may carry case reports and epidemiologic, methodologic and statistical information. Each audience is given a layer it can stop at, and the layers are ordered so that stopping early still leaves the reader with the action.
The register shifts deliberately between those layers. Persuasive language is used to convey new safety information to health care professionals, who will in turn advise or monitor patients, while patients are told to take specific actions to minimise harm and maximise benefit. The constraints the team works under are worth naming because they are not obvious: writing for multiple audiences, communicating uncertainty, and working within legal and regulatory limits — among them that the agency may not recommend one drug over another, because it must provide unbiased information.
The decision to publish at all sits inside a documented sequence. Early in development the experts discuss whether and when to communicate and which tool to use, examining the safety signal and reviewing additional data where needed; a communication may be issued before a drug is withdrawn from the market or its label changed, and the agency's own decision diagram runs from a potential safety signal through receipt and review of additional data to either market withdrawal or a labelling change, with a signal judged not credible closing the branch. In general a communication is issued for issues affecting large numbers of patients, potentially serious or life-threatening adverse events, new contraindications or warnings, a previously uncharacterised drug–drug or drug–disease interaction, or a medication error that could cause a serious or life-threatening reaction.
Most unusually for this genre, the effectiveness of the communication is measured, and the measures are reusable by anyone publishing safety information. The agency uses web analytics, media monitoring and public feedback: spread is measured as total web posts per day, with major dissemination occurring within one to two weeks of posting; diffusion is the breakdown of media types carrying the message, on the reasoning that diversity of media increases the likelihood a message will be adopted; and duration is the number of days the topic continues to generate web posts after launch. Focus groups solicit feedback from health care professionals, patients and other stakeholders. Adoption, spread and persistence are three different things, and a communication can succeed on one and fail on the others. The stated aim the whole apparatus serves is worth writing on the wall of any safety-communication team: that the right people get the right message at the right time.
What feeds the pipeline in the first place is the routine safety reporting system, which is the connection between this work and the reporting of harms: adverse events are reported to the agency on its voluntary and mandatory reporting forms and entered into its adverse event reporting database, and the agency's drug safety experts review that database to evaluate and assess signals of emerging risk, reviewing the medical literature for early adverse events alongside it.
Build trust before you need it, and expect the message to carry risk of its own
The vaccine-safety session gave three practices for establishing credibility, and each of them is a decision made before a problem arises rather than during one: announce problems early, explain both potential and actual risks, and include contrary information instead of withholding it. Withheld contrary evidence is the failure mode that converts a manageable safety question into a durable controversy, because the omission is discoverable and the discovery is what people remember.
The vaccine record supplies the illustration. Concern about thimerosal, a mercury-containing preservative, persisted long after in-depth studies of neurologic disorders and autism found no link, and one speaker reported that his website's inquiry statistics spiked with every burst of media activity on the subject even though the research had disproved it. Headlines linking vaccines to autism were followed by falls in coverage, and in the United Kingdom by a rise in measles cases which then exported — an outbreak of mumps in Iowa was traced to United Kingdom cases. A claim that has been answered can go on being asked, and communication planning has to budget for that rather than treat the evidence as closing the matter.
That session's most useful caution was that risk communication is not risk free: the public can respond in unforeseen or negative ways to what is communicated. The removal of thimerosal from most vaccines in response to public fear is the worked example — a precautionary action taken to reassure, after which autism rates continued to rise in the affected populations, so the action neither settled the question nor removed the concern. Anticipating how a message will be read, including by people who will read it as confirmation that there was something to fear, is part of drafting it rather than part of evaluating it afterwards.
The 2013 EMWA patient-information panel put the trust problem on the institutions rather than on the reader. Asked whether the pharmaceutical industry could be trusted to give correct information, its members answered that good medical writers are needed but that the problem may lie with overzealous marketing departments, and that a package insert has to consider patients' needs as well as the regulators'; their conclusion was that industry and clinicians must earn public trust by increasing their transparency. Asked whether times were changing with respect to transparency and firewalls, they agreed that they were, believing that the industry was trying to do the right thing and to be a responsible partner in healthcare, and that regulators were thinking of safety first. That is a panel of industry and clinical insiders assessing their own sector in 2013, and it is reported here as such.
Neff Rollins, at the 2009 risk-management session, framed the same need as a mandate for medical communicators to restore and build public trust, and pointed it back at the writer. Communicating candidly, clearly and accurately, in plain language wherever possible, assumes a communicator who has taken the trouble to become an informed consumer — taking courses to understand statistics and quantitative risk-benefit analysis, and finding out what the regulator does and does not do.
The regulator was working on the same problem in public. Prerna Mona Khanna, a physician consultant who also spoke at the 2009 risk-management session, described the FDA's Risk Communication Advisory Committee, formed in the wake of the Vioxx, Celebrex and Bextra episode to review and evaluate strategies and programmes for communicating with the public about the risks and benefits of the products the agency regulates. Her summary of what the agency had failed to convey is the useful part: its role is not to guarantee that an approved product is completely safe, but to determine that a given drug carries acceptable risks that are outweighed by its benefits. The committee met monthly through that year on improving public communication about the risks of regulated products, including both medical therapies and food safety, and a draft risk communication strategic plan was online at the time of the session.
What that 2009 panel and its audience made of the committee says something about where medical writers stood in the arrangement. The agency wanted to communicate directly with the public, presenting more science-based balanced risks and benefits rather than reaching people through the lens of sensational media, yet its annual science symposium on medical writing admitted only credentialed journalists; Grodberg's response was that the association needed to do better at making regulators aware of what professional medical communicators could contribute to public understanding of health risks. One attendee drew the parallel between the previous decade's effort to improve health literacy and a newer effort to improve risk literacy. The obstacles named for either effort were cultural, language and disability barriers, the media's predisposition toward sensationalist coverage, and the way the internet has changed where both consumers and physicians get their information.
Related topics
The numerical side of the same problem — absolute and relative risk, and how a treatment effect is expressed — is in risk and treatment effects. The vaccine science behind these examples is in antiviral drugs and vaccine platforms, and the wider question of how much a reader can do with health information at all is in readers who need plain language most.