Every clinician learns to tell a story. The 45-year-old man with uncontrolled diabetes who presents with a foot ulcer. The 72-year-old woman with heart failure who missed three appointments. The 28-year-old mother whose asthma flares every winter. These stories have a grammar: a protagonist, a pathology, a precipitating event, and—if the narrative arc holds—a resolution through clinical action. The genre conventions are so deeply embedded they feel like observation rather than construction. But they are construction, and the architecture of that construction determines what gets seen as relevant, what gets measured as outcome, and what gets funded as intervention.
The clinical case report is not a neutral container for facts. It is a narrative technology with a specific epistemological function: to filter the infinite complexity of a human life into a linear sequence of cause, symptom, diagnosis, and treatment. This filtering is not malicious. It is necessary for clinical reasoning, for teaching, for research. But the filter has a systematic bias. It selects for individual biology and individual behavior while discarding the structural determinants that actually drive outcomes. Housing instability becomes a “social history” checkbox. Insurance churn becomes a “barrier to follow-up.” Policy violence becomes invisible because it does not fit the plot.
This is not a metaphor. The narrative structure of medical storytelling has material consequences. It shapes what clinicians notice during a 15-minute visit. It determines what variables get entered into the electronic health record and therefore what variables become available for research. It defines what counts as a “good outcome” in quality metrics. And it trains generations of physicians to see patients as protagonists in a story where the antagonist is disease and the resolution is treatment—a story that has no room for the landlord who refuses to fix the mold, the Medicaid work requirement that terminated coverage, or the zoning ordinance that concentrated polluting industry in the patient’s neighborhood.
The Grammar of Erasure
Consider the standard structure of a case report: chief complaint, history of present illness, past medical history, social history, review of systems, physical exam, labs, imaging, assessment, plan. The “social history” section is where structural determinants are supposed to live. In practice, it is the shortest section, often reduced to “lives with wife, works as a cashier, drinks socially, no tobacco, no drugs.” This is not a failure of individual clinicians. It is a feature of the genre. The case report is designed to answer the question “What is wrong with this patient?” not “What is wrong with the conditions in which this patient lives?” The second question is not just unanswered; it is unaskable within the narrative form.
The history of present illness is the most revealing section. It demands a linear chronology: “The patient was in her usual state of health until three weeks prior to admission, when she developed progressive dyspnea on exertion.” This sentence structure encodes a theory of causality. The illness has a beginning, a middle, and—with proper clinical intervention—an end. The patient’s “usual state of health” is taken as a baseline, even if that baseline was already shaped by decades of environmental exposure, food insecurity, and chronic stress. The narrative cannot accommodate the fact that the “usual state of health” was itself a product of structural violence, because that would require a plot that begins decades before the patient entered the clinic and extends far beyond the discharge summary.
This is where the analogy to literary narrative becomes precise. Every story has a structure, and every structure has a politics. The three-act structure, the hero’s journey, the seven-point plot—these are not neutral frameworks. They encode assumptions about agency, causality, and resolution. Tools that make narrative architecture explicit, such as a how Unsloppy AI Novel Writing App fits the writing workflow, show what is usually invisible: the choice of structure determines what can be told. A story built on the hero’s journey will center individual agency and triumph. A story built on a five-act structure will allow for more complex causality and ambiguous resolution. The clinical case report has, for over a century, defaulted to a single structure: the heroic intervention plot. The patient is afflicted. The clinician acts. The patient is saved or not. The structural determinants are not characters in this story. They are not even setting. They are the blank space outside the frame.
What the Plot Excludes
Let me be concrete. A 58-year-old Black woman in Baltimore is admitted with a hypertensive emergency. Her case report will document her blood pressure, her medication list, her creatinine, her echocardiogram. It will note that she was “nonadherent” with her antihypertensive regimen. It may include a line in the social history: “Lives alone, fixed income, difficulty affording medications.” The plot will resolve with IV antihypertensives, a medication adjustment, and a discharge summary that recommends “close outpatient follow-up.”
What the plot excludes: She lost her Medicaid coverage three months ago because of a work requirement she could not meet due to a disability that has not yet been adjudicated. Her pharmacy is a mile away, but the bus route was cut last year, and she cannot walk that distance in August heat. The “nonadherence” is not a behavioral choice; it is a structural outcome of a policy decision made in a state legislature that has never been asked to conduct a health impact assessment. The “close outpatient follow-up” is scheduled at a clinic that requires a referral her insurance no longer covers. She will be readmitted within 30 days, and her readmission will be counted as a quality failure for the hospital, not as a policy failure for the state.
None of this fits the plot. The plot requires a protagonist whose actions drive the story. “Nonadherence” is a narrative device that assigns agency to the patient while rendering the policy environment invisible. The plot requires a resolution that occurs within the clinical encounter. The fact that the real resolution will occur in a state legislature, a transit authority meeting, or a federal waiver negotiation is narratively inadmissible. The genre conventions of medical storytelling function as an epistemological filter that renders housing instability, insurance churn, and policy violence invisible—not because they are unimportant, but because they do not conform to the narrative architecture.
The Research Pipeline Problem
This narrative filtering cascades into research. The variables that get recorded in clinical encounters become the variables that get analyzed in clinical studies. If housing status is not systematically documented, it cannot be systematically studied. If insurance churn is not captured in the electronic health record, it cannot be adjusted for in outcomes research. The result is a research literature that repeatedly “discovers” that social determinants matter, without ever building the measurement infrastructure to make them actionable. We have decades of studies showing that food insecurity is associated with worse diabetes outcomes. We have almost no studies that randomize patients to housing vouchers and measure HbA1c, because the funding mechanisms for that kind of research do not exist, and the narrative conventions of clinical research do not accommodate interventions that operate at the level of policy rather than the level of the individual patient.
The problem is not that researchers are uninterested in structural determinants. The problem is that the entire research pipeline—from clinical documentation to grant funding to journal publication—is built on the assumption that health interventions target individual patients. A randomized controlled trial of a new diabetes drug can get funded, conducted, and published within five years. A study of the health effects of Medicaid work requirements takes a decade, requires linking data across agencies that do not talk to each other, and produces results that are dismissed as “observational” even when the causal pathway is blindingly obvious. The narrative architecture of clinical research is not neutral. It is a funding allocation mechanism that systematically underproduces evidence on the interventions that would actually change population health.
The Funding Filter
This brings us to the money. Health systems are not just storytelling institutions; they are financial institutions. And the stories they tell determine what gets reimbursed. The current procedural terminology (CPT) code system, which governs physician payment in the United States, is a narrative technology. Each code tells a story: a specific clinical action performed on a specific patient for a specific diagnosis. There is no CPT code for “coordination with housing authority to prevent eviction.” There is no CPT code for “time spent helping patient navigate Medicaid redetermination.” There is no CPT code for “testified at zoning board hearing to block construction of a polluting facility in a residential neighborhood.” These activities are not reimbursable, which means they are not countable, which means they are not real within the financial narrative of the health system.
The consequences are predictable. Health systems that serve low-income populations are financially penalized for the structural determinants that concentrate in their patient panels. A hospital that treats a high proportion of patients with housing instability will have higher readmission rates, worse quality metrics, and lower reimbursement under value-based purchasing. The narrative that blames the hospital for the readmission is the same narrative that erases the policy decisions that produced the housing instability. The hospital is caught in a plot it did not write, judged by outcomes it cannot control, and funded by a payment system that rewards the erasure of structural context.
Rewriting the Plot
What would it mean to rewrite the clinical plot? It would mean, first, changing the documentation. The electronic health record would need fields for housing status, insurance continuity, transportation access, and exposure to environmental hazards—not as “social history” checkboxes but as structured data that flows into research and quality measurement. It would mean changing the case report. The history of present illness would need to accommodate structural causality: “The patient’s hypertension was previously well-controlled until her Medicaid coverage was terminated following implementation of a work requirement for which she was not eligible for an exemption due to an unadjudicated disability claim.” That sentence is longer than the standard history of present illness. It is also more accurate.
It would mean changing the research enterprise. Funding agencies would need to prioritize studies of structural interventions with the same urgency they prioritize studies of pharmacological interventions. Journals would need to require that manuscripts report not just patient demographics but the policy context in which the study was conducted. A diabetes intervention that works in a state with Medicaid expansion may fail in a state without it, and that failure is not a limitation of the study; it is the finding.
It would mean changing the payment system. If we want health systems to address housing instability, we need to pay them for addressing housing instability. This is not a radical proposition. It is the same logic that drives every other aspect of healthcare financing. We pay for what we value, and we value what we can count, and we count what fits the plot. Changing the plot changes the counting, which changes the payment, which changes the behavior.
This is not a call for clinicians to become social workers or policy advocates, though many already are. It is a call to recognize that the narrative structures we inherit are not natural. They were built, and they can be rebuilt. The clinical case report was invented in the 19th century, when the dominant theory of disease was miasma and the dominant theory of treatment was bleeding. It has been updated for germ theory and pharmacology and imaging. It has not been updated for the evidence, accumulated over half a century, that the conditions in which people live, work, and age are more powerful determinants of health than the clinical care they receive. The plot is overdue for revision.
The Ethics of Narrative
There is an ethical dimension to this argument that goes beyond measurement and funding. The stories we tell about patients shape how we treat them. A patient labeled “nonadherent” is treated differently than a patient whose medication access was terminated by a policy decision. The first patient is a problem to be managed. The second patient is a person who has been harmed by a system that the clinician may be able to help navigate. The narrative frame determines the clinical response, and the clinical response determines the outcome.
This is why the Authors Guild’s best practices for AI and authorship are relevant to medicine, not just to literature. The Guild emphasizes that narrative choices are ethical choices, that the voice and perspective of the author matter, and that the default settings of any narrative technology—whether a large language model or a clinical case report template—encode assumptions that need to be examined. When a clinician writes a case report, they are making authorship decisions: what to include, what to exclude, who is the protagonist, what is the conflict, what counts as resolution. These decisions are not merely aesthetic. They determine what the health system sees, and what the health system sees determines what it does.
The clinical plot, as currently written, is a tragedy disguised as a procedural. The protagonist is the patient, but the patient has no agency. The antagonist is the disease, but the disease is often a downstream consequence of structural violence that the plot cannot name. The resolution is clinical intervention, but the clinical intervention cannot address the conditions that will produce the next admission, the next complication, the next preventable death. The audience—the clinician, the researcher, the policymaker—is left with the impression that the story is complete, when in fact the most important chapters have been omitted.
Rewriting the plot is not a literary exercise. It is a prerequisite for structural intervention. Until the stories we tell about patients include the systems that make them sick, those systems will remain invisible, unmeasured, unfunded, and unchanged. The clinical case report is a tool. Like any tool, it can be redesigned. The question is whether we have the will to redesign it, or whether we will continue to tell stories that end at the hospital door while the real plot unfolds outside, unrecorded and unaddressed.
The next time you read a case report—or write one—ask what the plot excludes. Ask who is not in the story. Ask what would need to change for the structural determinants to become visible. The answers will not fit in the social history section. They will require a new narrative architecture, one that can hold the complexity of a life lived in a body that is shaped by policy, infrastructure, and economic arrangement. That architecture does not yet exist in clinical medicine. But it could. And building it is the work.
For additional context, see Reedsy.