In a workshop on narrative in public health in Lomé, Togo, I ask participants—ministry of health workers and representatives from non-governmental agencies—to share examples of stories they use in their work.1 These include theatre pieces and personal illness narratives employed for community health promotion, examples which reflect the long-standing use, worldwide, of narrative to promote health. Indeed, a growing body of research illuminates the power of stories to influence health behaviour.2–5 When asked what stories they tell ministers and other people in power about the health consequences of the policies they enact, they grow silent.

One man offers, ‘We give them numbers’.

I ask them then to spend a few minutes thinking of events and people they encounter in their work that might put a face on the numbers for ministers and other government employees. When I call for examples after that, many hands fly into the air.

A man who had earlier talked about his work to improve young women's access to reproductive care tells a story about a girl who received a sewing machine as part of her employment programme when she ‘fell pregnant’. She died following an unsafe abortion. He describes her and even the sewing machine in great detail. She was 19 years old and though she might have been able to care for a baby, she was terrified of losing her opportunity to become a seamstress. Deaths resulting from unsafe abortion are common in the developing world and the highest rates are in sub-Saharan Africa6 but in one vivid mention of a sewing machine, this particular young woman moved out of the realm of statistical abstraction for the people in that room.

During a similar workshop with students in my Master of Public Health programme, I ask them to write an account of an incident that represents public health. They often write about the very moment they decided to enter the field. A student studying rural-urban health disparities writes about working as an emergency medical technician in rural Missouri and watching the victim of a motorcycle accident die on the way to the hospital. A nurse from Haiti writes of the ‘awkward patchwork design made of asphalt, clay and mud’7 on the road on her way to a clinic where she conducted health education for sex workers, along with the story of a woman paid to have intercourse in front of her own children.

When asked where she wants to focus in public health, this nurse will tell you, ‘the social determinants of health’. When asked why, she will tell you about that woman in Port au Prince.

Humanities in health has until now been primarily understood to mean humanities in medicine and has generally failed to include public health. I will argue in this paper that the common justifications for the former—including increased empathy among practitioners—are at least as applicable, if not more, to the latter. There are numerous and growing examples of writing workshops in medical schools and even, at the Medical Center at Columbia University in New York, a graduate degree in narrative medicine.8–11 Many programmes seek to provide doctors with a creative outlet for exploring the challenges of the role through writing. Most incorporate the study of literature into medical training with the intention of improving the clinical encounter by encouraging empathy in doctors, nurses and others involved in treating patients.9 ,10 ,12

Developing empathy for patients involves cultivating ways to understand them as individuals who are more complex than their presenting illness or condition.13 Advocates of humanities in medicine have long argued for a practice of medicine that includes attention to the social, family and cultural life of patients.13–15 The growing burden of chronic disease, as well as shifting models in healthcare financing, further raise the expectation that the training of clinicians will include cultural competence and awareness of health equity issues.16 ,17

As it does, the practice of medicine and the training of physicians begins to overlap in important ways with public health education and practice, which makes the absence of humanities programmes in the latter all the more notable. While cultural and ethical awareness is an explicit aim of what Charon refers to as ‘narrative competence’ in medicine, these are qualities no less necessary in public health, which requires professionals who can engage with communities and across disciplines in a culturally competent manner.10 ,18 ,19 The emphasis on interdisciplinary work in public health, as well as the role of public health professionals in translating epidemiological evidence into social policy,20 further argues for the potential role in public health of two elements of humanities in medicine programmes in particular: creative non-fiction and fiction writing.

‘Lies, damned lies and statistics’: creative non-fiction in support of sound policy

Storytelling is not an obvious topic for public health curricula for a number of reasons.21 The fact that public health is prevention focused poses a familiar dilemma: how do you describe a disease that wasn't spread or an injury that did not occur? This is a challenge that Psychology Professor Paul Bloom highlights in his essay in The New Yorker entitled ‘The baby in the well: the case against empathy’. Bloom argues that vivid stories undermine sensible social policy precisely because the most interesting ones illustrate unusual events rather than representative ones. He argues that ‘representative stories fade into the background like the hum of traffic’ and ‘sensible policies often have benefits that are merely statistical’.22 Thus, stories steer us towards policy responses that are ‘parochial, narrow-minded and innumerate’.22

When Dr Bloom warns that, ‘empathy will have to yield to reason if humanity is to have a future’,22 he would likely find many graduate public health students who agree. Students study public health to tackle the most real of issues, after all: lives lost to preventable diseases, inadequate access to care, disparities in health connected to economic class, race, and other social determinants. Their first in-depth study of the epidemiology behind these phenomena is powerful; seeing patterns where none were previously visible can feel like seeing clearly for the first time.

When you look at events through an epidemiological lens, it quickly becomes clear that things you might have thought of as random occurrences or the result of bad luck are predictable, quantifiable and, if society acts wisely, preventable. If you crash your car at a busy intersection, the story you're likely to tell is about how you were momentarily distracted by your kids’ argument, or about how a white Toyota seemed to come out of nowhere. Either of those things may be true, but once you discover that the intersection where you crashed has the highest number of crashes in the city and that minor engineering changes have been shown to reduce crashes at similar intersections by 50%, the meaning of the story changes. The bright light of correlation and significance illuminates previously murky relationships between risk and outcome. It's no wonder that the anecdote seems to lose its power in comparison with those numbers. When public health students see policy-makers tune out data in favour of anecdotes that contradict the evidence, they often think that the answer has got to be more evidence.

Furthermore, unlike clinical medicine, population health approaches the subject of human experience from the aggregate, rather than the individual. There is an old adage in creative writing that says if you want to write about mankind, write about one man. A story about shared human experience in the abstract is boring. One must begin at the micro—the lived experience of a singular individual—and in the micro find the macro—universal truths about what it's like to be a human being at this or any other time.

By focusing on communities rather than individuals, the study of epidemiology searches for patterns like a fire spotter in the sky rather than a fire fighter on the ground. The emphasis is not on the contours of a particular mountain, but on the way one mountain links to another in a range and flattens over distance into the plains. One woman's decision not to use a condom during sex becomes part of a pattern that includes many women, as well as factors like gender discrimination and education level and income. What's unique about the first woman's situation, her individual story, can actually be distracting when the goal is understanding patterns that affect many individuals at once.

Former legislator John E McDonough reminds us, however, that human beings, including policy makers and scientists, ‘live our lives crafting, telling and receiving stories’.23 Though the field of public health is evidence-based, we forget at our peril that the legal definition from which the very concept of evidence is derived includes personal accounts of events.20 If what McDonough further claims is true and ‘stories can enable law-makers to understand a legitimate need for policy change but just as readily can lead them to make bad policy decisions,’23 why is so little, if any, time devoted to learning how to tell stories, and how to tell them effectively, in public health programmes?

Since no legislator wants to craft laws tailored to each individual's specific circumstances, population-based evidence should be ideally suited to policy making. So the question is, why do debates about public health policy so often lurch into unscientific territory, reflecting political concerns that apply evidence selectively or contradict existing evidence altogether?24 ,25 One possible explanation concerns the infrequent contact between academic researchers and policy makers.20 Another is that that when it is presented, statistical evidence rarely incorporates local data or cost-benefit analysis or is otherwise framed in a manner designed to support policy making.20 It may also be that our need for stories goes beyond entertainment and that data alone, no matter how significant, is not as persuasive as we think it ought to be.

Writing in the Journal of Economic Behaviour and Organisation, psychologist Robyn M Dawes argued that the ability of human beings to understand information on probability—and behave in the way that economists predict they should when faced with information about base rates—was directly dependent upon whether or not the information was attached to a causal story that made sense of the events.26 We may, in fact, be hard-wired to need anecdotes in order to absorb even the most statistically solid evidence. Indeed, research indicates that the most persuasive kind of message combines narrative and quantitative evidence.27

I would counter Bloom's claim that representative stories are inherently weaker with the argument that they are often poorly told. If presented in a way that illuminates and connects, familiar things can be just as compelling as unfamiliar ones. An important element of good writing, in fact, is the ability to conjure in the reader the satisfying feeling of recognition, the felt response, Yes, that's exactly the way it is! The former director of the Iowa Writers’ Workshop, Frank Conroy, used to speak about the desire to read fiction as the longing to be accompanied. We read alone but we live with others; writing is a solitary form of labour and it is a deep form of human communication.

A story is uninteresting when it fails to be particular, not because it is representative. There is nothing unique about falling in love, yet countless novels are written about just that. The good ones inspire a ping of recognition in us, no matter how different the characters are from ourselves. Stories that illustrate what is likely, representative and widespread can be as compelling as those depicting rare events—precisely because of the feeling of recognition they inspire. The prevailing definition of public health includes ‘organised community action’28 and a key catalyst for organising is understanding that the situation you find yourself in is not unique. For policy-makers, well-told stories can provide an ‘anchor for statistical evidence’.20 For community members, they can help the individual find herself within larger social phenomena to great personal and political effect.

Investigations by Niederdeppe et al29 into attribution theory further suggest that narrative may offer more than just a tool to persuade, it may even perform a catalysing effect on one's willingness to make the argument in the first place. In research involving attribution theory, whereby people emphasise either ‘external’ or ‘internal’ factors in their understanding of causation, they demonstrate that stories that highlight social factors in health outcomes can in fact predispose many listeners to support policy interventions that address social causes.29 I'm not suggesting that public health advocates abandon the evidence; quite the opposite. I'm suggesting we identify the tools that will allow us to move beyond ‘pale statistical abstractions’ to create narratives that illustrate the data and move us to action.22

‘Not real but true’: fiction writing as a tool to develop empathy and cultural competence

A colleague in my Master of Public Health programme complained recently about the lack of understanding students displayed when discussing an article about smoking cessation strategies among poor women in Chicago.30 She teaches the core course Social and Behavioural Science in Public Health and while it seemed that the students grasped intellectually the principle that poverty affects health status, their ideas about changing behaviour all focused on educating the women in question about the dangers of smoking, without revealing any real understanding or empathy with the day-to-day pressures the women faced.

She lamented, ‘I wish there was a way to make these contextual issues come alive for them’.

American students of public health share with Americans, in general, the overall tendency to attribute health outcomes with behavioural risk factors to internal causes rather than societal factors.29 These beliefs coexist uneasily with what they learn—and intellectually grasp—about the social determinants of health. In my experience, and in my colleague's, when the task of imagining what it's like to make health decisions amid social and economic constraints proves too difficult, students fall back on an internal narrative of which they are often not even aware: that health outcomes are predominantly the result of personal choices and that the task of health professionals is to convince people to choose otherwise, regardless of context. The reading and writing of literary fiction holds promise, in my view, for the training of health professionals who can, as called for in the Institute of Medicine report Who Will Keep the Public Healthy, ‘look beyond the biological risk factors that affect health and seek to also understand the impact on health of environmental, social and behavioural factors’.31

David Comer Kidd and Emanuele Castano are among the social scientists who have found that reading literature is an effective tool in developing empathy and social intelligence. Unlike genre fiction which emphasises plot, literary fiction focuses on character and engages readers in imagining the ‘circumstances and inner lives of people unlike themselves’ and in considering ‘multiple points of view’.32 The theoretical framework underlying their investigations involves the power of literature to improve affective Theory of Mind, which is closely linked to empathy. Exposure to fiction that makes demands on readers to think deeply about character appears to help us recognise our similarity with people who initially seem unlike ourselves.32

Research by Murphy et al33 suggests that emotional involvement and identification with fictional characters featured in narrative can augment changes in knowledge, attitudes and health-related behaviour intentions. For purposes of health behaviour change, this effect was found to be most powerful when participants also identified personally with the characters in the story.33 But even when participants did not consider themselves personally similar to the characters in the narrative, the potential of stories to inspire empathy is enhanced when the experience of the narrative promotes their ability to ‘adopt the character's point of view’.

Tal-Or and Cohen draw further distinctions between two commonly referenced theories of human involvement in stories, both of which have implications for the potential of fiction writing to enhance empathy and cultural competence among public health students: transportation and identification theory. Transportation involves a strong identification with the narrative itself (the story) but does not specify with which element the reader is engaged.34 Identification, on the other hand, represents a strong connection with a specific character.34 What follows is an example from my public health capstone of how engaging students in the writing of literary fiction focused deliberately on character development rather than story might enhance their ability to empathise with individuals unlike themselves.

Each student is asked to write down a health behaviour that they find particularly difficult to understand and empathise with. They are then asked to take a mental snapshot of an individual engaging in that behaviour. On a separate sheet of paper they are asked to describe the person in that snapshot according to demographic characteristics commonly used in public health analysis: age, gender, racial or ethnic background, highest level of education attained, rural or urban status, sexual orientation and economic status.

This is where a lot of public health analysis stops. It is also the place where, once again, individual and collective experience intersects and the potential of literary fiction to help public health professionals balance effective understanding of both can be illuminated. Rather than stopping with the demographic profile, students are introduced to the concept of creating round characters in literary fiction. They are asked to give their character a name and answer a variety of questions designed to move these initially flat characters in the direction of ‘complicated individuals whose inner lives are rarely easily discerned but warrant further investigation’.32 In my workshop, these include questions such as, ‘What is your character's dream job?’ and ‘When your character walks into a room full of strangers, the first thing she thinks is…’

Exploring a character's motivation is a central focus of literary writing and a reductionist approach—boiling that motivation down to a simple explanation, whether external (membership in an ethnic group or gender) or internal (ignorance or paranoia) results in clichéd, uninteresting work. In creating scenes around a rounded character—a human being—rather than a concept, students are compelled to consider the impact of social determinants on individuals, not groups, and to flesh out that character's world with real, familiar and believable elements. This approach has the potential to address a common pitfall among diversity curricula: teaching strategies based on stereotyped behaviour of identified cultural groups.35 ,36

As the exercise progresses, students write two short scenes involving their character: one in which she is on her way to work and the second in which she is engaged in a difficult task with another person. Discussion questions at each phase prompt participants to explore the extent to which their characters are built on or begin to transcend stereotypes. As do most writers of character-based fiction, student participants generally find that their understanding of the character deepens through this imaginative work and that their feelings towards the character—and even towards the causes of the triggering behaviour—begin to change. Some sample comments: I really disliked this character when I was first describing her, but now I'm starting to like her better. At first I thought that the character wasn't smart, but then I realized he simply had not attended college. It reminded me that there are a lot of reasons someone might not attend college and that not everybody who fails to go is unintelligent. I feel like I understand why she did it.

The open-endedness of the exercise and the element in which the author's own imagination is implicated in exploring the complexities of health behaviour has potential, also, to address one of the hazards of narrative raised by Angela Woods.37 The imposition of narrative, she argues, is not inherently a good thing in and of itself. She draws attention to the research of Yiannis Gabriel who notes, ‘while stories can be vehicles of contestation, opposition and self-empowerment they can also act as vehicles of oppression, self-delusions and dissimulation’.38

In a sense, the exercise described above uses a narrative technique (building character) to subvert an inner narrative (attribution of behaviour). The students are called upon to imagine deeply the lives of their characters in order to challenge the stories they tell themselves about people most likely to engage in the behaviours they seek to change, stories they may not even be aware of themselves.

In the writers’ toolbox

In discussing the use of fictional narratives to address disparities in cervical cancer screening among Latina populations, Murphy et al33 point out that ‘narratives may not be as easy to construct and produce as non-narratives’ and suggest that health communication researchers ‘reach out to professional storytellers’ to enhance the quality of their stories. While not discounting the production power of media professionals, I propose that some training in storytelling technique among those in public health might go a long way towards addressing the demand that we become, if not someone with ‘Mark Twain's narrative ability’, then at least people who recognise the elements of stories and have some skill in constructing them.

Of the many narrative truths that public health professionals might learn from fiction writers, one of the most profound is that one's own desire to tell a story does not ensure that anyone else wants to hear it. The pressure on a fiction writer is obvious from the first sentence. Why should anyone care about your made-up character in your made-up town with your made-up conflicts? Unless you get a reader to care right away she will put the book down and not pick it up again. The stories that public health professionals tell to influence policy should be based on epidemiological evidence and a thorough understanding of the policy process. Many would argue that accurate representations of important public policy matters supported by convincing data should not require the help of narrative craft. But I'll put that should against the stacks of evidence overlooked by policy-makers and the public and we can guess which will win.

In Nobel Laureate Alice Munro's novel Lives of Girls and Women a character states, “The main thing was that it seemed true to me, not real but true, as if I had discovered, not made up, such people and such a story.”30 Sensory details are one of the tools that can render a fictional story easier to believe—can make it seem more true—than one that is based on the historical record. On any page of a literary work you will find recognisable details about familiar objects and experiences: peeling an orange, holding the hand of a child, being punched in the stomach. No matter that you will have likely experienced these things yourself, a good writer knows how to create a scene and pull you into it. You will be able to almost taste the tangy smell of the orange's peel as it is broken, feel the stickiness of the child's palm, or follow the pain of the punch as it leaps from the stomach to the throat. What follows are some additional elements that the craft of literary writing can offer in creating narrative skills among public health professionals.

• Voice: Sometimes confused with message or point of view, a writer's voice is created through a coherent and deliberate set of choices in tone, language, ordering of scene, and perspective that give the reader the sense that another human consciousness is behind a given piece of writing and guiding the reader through it.

• A balance of description and scene: Compelling writers learn how to show and not tell, to employ scene and dialogue to bring the reader into the story. Vivid language and imagery and dialogue that ‘speaks for itself’ create the sense in the reader that she is present to experience the action, rather than merely hearing about it secondhand. Why is it important for a writer to bring the reader inside the story? Because the purpose of good writing goes beyond informing the reader; the goal is to move her. An emotional connection to the content depends on bringing the materials close to the reader and the reader inside the world that is being written about.

• Urgency: Writing instructors will often ask, ‘What is at stake?’ when the urgency behind a story hasn't yet made it sufficiently to the page. The stakes in public health could not be higher; but that urgency does not make a case for itself any more than statistics alone tell a generally persuasive story. Urgency can be conveyed through narrative tools such as active language and narrative drive—the sense that the piece has strong forward motion.

• Non-technical language: In A New Way to Talk About the Social Determinants of Health, the Robert Wood Johnson Foundation observes that, ‘when messages are presented in colloquial, values-driven, emotionally compelling language, they are more effective’.39 A focus on the craft of literary writing gives advocates well-needed practice in using vivid, clear and accessible language.

• Well-chosen detail: While well-chosen details can bring a reader into the moment that is being described in a piece of writing, extraneous and unnecessary ones can just as easily distract the reader from the central focus of the piece, and send them down unrelated rabbit holes in search of meaning. Learning how to choose the right details—the ones that make a scene more vivid and convey important information that will enhance the reader's understanding of the characters and setting—is, like the other elements of the literary writers’ craft, a skill that improves with practice.