Many Patients With CKD Never Receive a Diagnosis

LONG BEACH, Calif. -- Despite seeing a healthcare provider, nearly a third of individuals with chronic kidney disease (CKD) went undiagnosed, researchers uncovered.

Roughly 2.85 million patients with a qualifying hospital visit were identified with either clinical CKD or at risk for CKD. But of 1.4 million patients that met the clinical criteria for CKD, 31% were undiagnosed, Joel Topf, MD, medical director of St. Clair Nephrology Research in Detroit, and colleagues reported.

This represented 440,749 individuals who met Kidney Disease Improving Global Outcomes (KDIGO) clinical CKD criteria, involving at least two test results 3 or more months apart with an estimated glomerular filtration rate (eGFR) under 60 ml/min/1.73m² or urine albumin-to-creatinine ratio (UACR) over 30 mg/g, according to a poster at the National Kidney Foundation (NKF) Spring Clinical meeting.

"Thinking about 440,000 people who are engaged in the medical system but are being failed by that system is upsetting. That's a lot of moms, dads, aunts, and other loved ones not getting the care they should," Topf told MedPage Today.

"This is not just unselected Americans, but people that are already engaged in the health system. These are people that are expecting and deserve quality care. These people should know what diseases they have," he continued, saying he was "disappointed" by this finding. "Our medical system is letting people down. The diagnosis of CKD can't be made in a single point of time, it requires longitudinal follow-up. It's right in the name, it's chronic. But what we found was that this 'chronicity' can be established by grabbing older data from a prior encounter."

"I personally assumed that these undiagnosed CKD were primarily in people that were not engaged in the medical system," Topf added. "But here we found people who were getting tested and seeing doctors and were still leaving visits with an unnecessarily incomplete view of their health."

These findings point to "an incredible amount of inefficiency where providers are building a patient's health database," he added, pointing out the database suggests the diagnosis, but then no one is closing the loop and giving it to the patient.

"These inefficiencies now matter more than ever because we have drugs and therapies that can slow the progression of CKD, but these therapies don't cure CKD. If they did, then getting started early wouldn't matter. But since the therapies only slow the progression, starting as early as possible is essential to prevent severe disease, dialysis, and transplantation," he said.

Most of the undiagnosed CKD patients had at least one comorbidity; hypertension was the most common (67%), followed by hyperlipidemia (46%), and diabetes (31%). Most patients with undiagnosed CKD had stage 3 (82%). Compared with CKD patients with a diagnosis, those who went undiagnosed were more often female (63% vs 45%) and white (78% vs 72%).

Undiagnosed patients also had a high frequency of healthcare involvement. Forty-three percent had at least one emergency department visit within the year, while 26% and 10% had a hospitalization and intensive care unit admission, respectively.

For these undiagnosed patients, there was an estimated total all-cause annual healthcare cost of $6.7 billion. In 2021, average annual all-cause healthcare cost was $15,196 per undiagnosed CKD individual, which increased with CKD stage:

• Stage 3: $14,403
• Stage 4: $19,136
• Stage 5: $21,081

For the study, researchers utilized the PINC AI Healthcare Database representing data from over 1,200 hospitals. Adults with inpatient or outpatient visits from 2017 to 2021 and ≥1 serum creatinine with eGFR <60 mL/min/1.73m2 or UACR > 30 mg/g were included. Patients visiting hospitals without continuous data submissions, or those with a history of extracorporeal membrane oxygenation procedures or kidney transplant were excluded.

Of the 2.85 million patients included, half (1.42 million) were identified as being at-risk for CKD. These individuals had at least one test result with an eGFR under 60 ml/min/1.73m² or UACR over 30, but did not meet KDIGO criteria.

"These are people who have half a diagnosis of CKD," Topf pointed out. "Now perhaps that albuminuria was transient due to a fever or exercise, and maybe that GFR was spurious or was due to [acute kidney injury]. And I'm sure that accounts for some of that 1.4 million, but for the rest they actually have CKD and they are just waiting for the confirmatory follow up 3 months after the index study."

"I had never thought about this group and it is pretty stunning that it is so large," he said. "In the most recent CKD Guidelines from KDIGO, they recognize and talk about this group and actually recommend physicians consider treating them. Pretty interesting when you think about that especially given what a large fraction of the people we looked at fell into this bucket."

While these findings highlight there are quite a few patients "floating around without a diagnosis," Topf said he doesn't think this is something that individual doctors can solve, and instead needs to be tackled by policymakers and people with a more strategic view of the problem.

"When you find that a third of people meeting the criteria for CKD are not being diagnosed, this reflects system-wide failures and we need a systematic approach to gathering disparate data and pushing this to physicians. We can't have doctors going on an information scavenger hunt to make a diagnosis," he said. "The promise of electronic health records and artificial intelligence is to do this type of work for us, so doctors can spend time listening to and building trust with their patients."

Comment