American health systems struggle consistently with a particular category of patient, for reasons that have very little to do with clinical competence.
The patient I have in mind has a documented finding of some kind: an aneurysm measured below the diameter that warrants repair, coronary calcification without symptoms, a nodule that calls for an interval scan rather than a biopsy.
Correct management in each of these cases is an opportunity for preventive strategies and requires surveillance across a period of years, which depends entirely on somebody remembering to deliver it. Every health system in the country carries thousands of such patients, and very few could produce a list of them on request.
Health systems are organized around episodes and thresholds for reasons that are entirely rational. Reimbursement pays for encounters and procedures, quality measurement counts events and outcomes attached to a defined episode, and specialty scope tends to be defined by whatever a given specialist intervenes on.
Anyone who has spent time on the administrative side of medicine understands how completely those forces shape an institution; working together across several decades, they have produced organizations that perform impressively the moment a patient crosses a line.
However, a patient who has not crossed that line falls into the space between the categories, unclaimed by any specialist, often, because there is nothing yet to intervene on, held nominally by a primary care physician with eleven other items to address in a fifteen-minute visit, and released by an imaging department that completed its work correctly and moved on to the next study.
Organizations perform impressively the moment a patient crosses a line.
Until then, a patient is adrift in the gaps between care.
Trace such a patient backward and what emerges is a sequence of sound decisions. The radiologist described the finding accurately, the specialist applied the appropriate threshold, and the primary care physician triaged reasonably against everything else competing for attention that morning.
Every one of those judgments holds up on review, which is precisely why the problem resists solutions built on exhortation, education, or asking people to try harder.
We use the phrase watchful waiting as though the waiting were the demanding part. Watching is the demanding part, and at scale it becomes a question of organizational capability rather than clinical judgment.
Following one patient for eight years means holding an interval in mind while they move among a primary care practice, multiple imaging locations, and a specialist or two’s office. It means that commitment must continue after the departure of the coordinator who tracked it in a spreadsheet, a change of insurance, and the year they skipped their physical altogether.
Above all, it means someone noticing absence, which is considerably harder than noticing a finding, since nothing arrives in an inbox when a patient simply stops coming back.
Three structural deficits show up wherever I look at this.
Responsibility is spread thinly across several roles, and responsibility distributed that widely tends to be exercised by none of them once volume climbs past a certain point.
Manual surveillance performs adequately through a pilot and degrades as the panel grows, so organizations generally locate the ceiling well after they have committed to the program.
Patients move among sites, systems, and payers, leaving individual findings scattered across records that do not speak to one another. What disappears is the thread connecting this year's finding to next year's obligation.
Several trends are enlarging this category simultaneously. Imaging volume keeps rising cross-sectional imaging detects far more incidental disease than it did a decade ago, the population is aging, and screening programs continue to expand. An organization holding its follow-up capacity constant is therefore falling behind by default.
The strategic consequences follow from the clinical ones. When patients with documented disease have no continuity of care, they receive their eventual treatment wherever they happen to land, which may or may not be the system that found the disease in the first place. A documented finding with no plan attached to it sits in the chart as an exposure. And arriving at intervention later than necessary costs more and goes worse, for the patient and for the institution alike.
Closing this gap requires infrastructure, which is a difficult thing for leaders to hear. Effort is what we know how to ask for; infrastructure has to be funded, staffed, and sustained long after the enthusiasm of the launch has worn off.
In practice, ownership means naming a responsible party for every patient under surveillance and specifying the interval instead of assuming it. The tracking has to survive staff turnover and patient movement, and it has to flag absence rather than waiting on a visit that may never be scheduled. Measurement matters as much: an organization unable to report how many below-threshold patients it carries, and how many are current, has no way of knowing whether any of this is working.
The clinical medicine here was settled years ago. What remains is organizational work, and it belongs to those of us who build and run these systems.
Learn more about Eon’s Cardiovascular solutions.
_____________________________________________________
Common Questions
It’s highly variable and often informal. Primary care holds the patient nominally, but a provider with numerous other issues to cover in a short visit isn't positioned to track an interval that runs for years. Ownership means naming a specific responsible party for every patient under surveillance and specifying the interval aligned to the patient’s risk.
It varies by finding, but multi-year tracking is common — an aneurysm under the surgical threshold, for example, can require monitoring across a decade. That span is precisely why manual tracking breaks down: it has to endure staff and insurance changes, patient care gaps, and it has to escalate automatically, before the patient stops returning for care.
The finding is documented, but no action is taken to monitor that patient. They may eventually get treated somewhere, but often not at the system that found the disease in the first place, and by the time it's caught, intervention tends to be more risky and costly.
Both, but staffing alone doesn't scale. A manual process can track a small cohort of a few dozen patients. It usually can't track the hundreds or more that most systems are actually documenting, especially as imaging volume and screening programs keep expanding. Closing the gap takes operational infrastructure that reduces manual work and extends care team capacity.
.png)