Why data and systems need to converge to help CKD patients
Kidney failure shouldn’t be the point at which care finally becomes coordinated. There’s ample lead time to act earlier and more effectively.
For many patients with chronic kidney disease, the healthcare system has the information it needs long before kidney failure occurs.
A declining eGFR is visible. Albuminuria can be measured. Diabetes and hypertension are known risk factors. A nephrology referral can be made. Education can begin. Transplant readiness can be discussed. Dialysis access can be planned.
Yet too often, these steps occur as separate events rather than as one coordinated journey. That is the real opportunity in kidney care.
We have spent years improving individual interventions across the chronic kidney disease continuum. The next step is connecting them so that one clinical signal reliably triggers the next appropriate action before the patient reaches a crisis.
CKD gives us an unusually long runway
Kidney disease is not a rare problem. The Centers for Disease Control and Prevention estimates that more than one in seven U.S. adults has chronic kidney disease, and as many as nine in 10 adults with CKD do not know they have it.
That combination — high prevalence and low awareness — should make CKD an obvious population-health priority.
Unlike many acute conditions, CKD often gives health systems years of usable data before kidney failure. Two of the most important markers are estimated glomerular filtration rate and urine albumin-to-creatinine ratio.
The problem is not that those data elements are exotic. The problem is what happens after they appear.
Does an abnormal result generate the right follow-up? Is CKD placed on the problem list? Does the primary care clinician know when specialist input is warranted? Can the nephrologist see the longitudinal trend? Is the patient educated while there is still time to make decisions rather than being introduced to dialysis during an admission?
That is where kidney care becomes a health data management problem.
A referral is not just a referral
One of the most consequential transitions in the CKD journey is from primary care to nephrology.
The current KDIGO CKD guideline emphasizes risk assessment and earlier management rather than waiting for a single laboratory threshold to force action. There is good reason to pay attention to timing.
A 2025 systematic review and meta-analysis involving more than 630,000 patients found that earlier nephrology referral was associated with lower mortality, fewer emergency dialysis starts, more permanent dialysis access and higher rates of kidney transplantation.
But “make a nephrology referral” is still too narrow a solution. Referral needs to become part of a managed pathway.
A patient with progressive CKD may need medication management, nutritional support, anemia management, cardiovascular risk reduction, social work, modality education, transplant preparation and eventually vascular or peritoneal dialysis access. No one of those functions is enough by itself.
The system should know what comes next
A mature kidney-care model should be able to recognize progression and create a predictable sequence of action. That does not mean every patient follows the same pathway. It means the system should know when the next conversation ought to happen. For example, when kidney function is deteriorating, patient education should begin before the word “dialysis” becomes an emergency.
When kidney failure becomes foreseeable, the patient should understand the differences among hemodialysis, peritoneal dialysis, transplantation and conservative management.
When transplantation is appropriate, evaluation should begin early enough for the patient to complete the workup and potentially explore living donation. KDIGO recommends referring potential transplant candidates six to 12 months before anticipated dialysis when possible, specifically to create time for evaluation and possible preemptive transplantation.
And if dialysis is likely, access planning should not wait until the patient is in the hospital. NIDDK advises establishing vascular access well before hemodialysis begins.
These should not be disconnected checkboxes. They should function as a coordinated progression model.
Crossing organizational boundaries
This is where many kidney-care programs struggle. Primary care may live inside one health system. The nephrologist may practice independently. Another organization may deliver dialysis. The transplant center may be miles away and on a different EHR. The insurer may have its own care-management program. Community organizations may address transportation, food insecurity or medication affordability without visibility into the rest of the care plan.
The patient becomes the integration layer. We should not expect someone managing advanced kidney disease to carry every referral, medication list, laboratory trend and care-plan decision from organization to organization.
The infrastructure should do more of that work. Health systems need longitudinal kidney registries that can identify progression, surface gaps in care and follow patients beyond the walls of the hospital. Referral status needs to be visible. Transplant evaluation should not disappear into a fax workflow. Dialysis access planning should be trackable. Hospital discharges should flow back to nephrology and dialysis teams rapidly enough to change the next treatment.
The technology exists to do much of this. What is often missing is ownership.
CMS tests the concept
Federal payment policy is increasingly acknowledging the same structural problem.
The Kidney Care Choices Model of the Centers for Medicare & Medicaid Services brings nephrologists and other kidney-care providers together around patients with late-stage CKD, kidney failure and transplantation. The premise is straightforward. Instead of waiting until a patient begins dialysis and then coordinating care, create accountability earlier in the disease course.
Early results offer both encouragement and caution. CMS reports improvement in measures such as optimal dialysis starts, home dialysis and pre-emptive and living-donor transplantation. At the same time, the model has not yet generated net Medicare savings. That distinction matters.
Better coordination should not be sold as a magic cost-reduction strategy. Its first obligation is to improve the patient journey and clinical outcomes. Sustainable economics have to be demonstrated parallel to that.
A longitudinal operating model
The kidney-care continuum gives us a clear test of whether healthcare is serious about using data as an institutional asset.
We can identify risk; we can see progression; we know many of the interventions that should follow. What’s left is connecting those signals across time, specialties and organizations so that patients do not have to reach kidney failure before the system finally behaves like one system.
The goal should not be to coordinate dialysis better. Rather, the target should be to coordinate kidney care early enough so that every patient reaches the next stage informed, prepared and with as many options still available as possible. That is a much more ambitious standard. It is also the one our data should finally make possible.
Zachary W. Sutton, DHA, MS, MSPAS, DFAAPA, FACHDM, is director of medical operations and transplant program coordinator for Carolina Nephrology.