Assessing the Impact of Reporting eGFR
Nephrology Visits Increased, But Were They Necessary?
By Genna Rollins
For nearly a decade, professional guidelines have defined chronic kidney disease based on glomerular filtration rate (GFR). However, even after various formulas for estimating GFR have been developed and eGFR is widely reported in the U.S. and elsewhere, little is known about the impact of reporting eGFR on nephrology visits and utilization of healthcare resources. This issue of Strategies examines a study that sought to look at these relationships.
In 2002, the Kidney Disease Outcomes Quality Initiative (KDOQI) of the National Kidney Foundation proposed a new definition and staging of chronic kidney disease (CKD) based on glomerular filtration rate (GFR) values. KDOQI did so because previously there had not been a uniform classification of CKD, and GFR had become widely accepted as a very good overall measure of kidney function in health and disease. KDOQI also suggested that reporting GFR would improve communication between patients and providers, enhance public education, promote dissemination of research results, and facilitate early recognition of and treatment for CKD. In the intervening years, labs in the U.S. and many other countries have begun routinely reporting estimated GFR (eGFR) based on several equations, and reporting is even mandated in some states. However, little is known about the impact of reporting eGFR. A recent study sought to examine this issue (JAMA 2010; 303:1151-58).
“We wanted to evaluate the impact on healthcare delivery of implementing eGFR reporting throughout the province—the number of visits, patient care, and outcomes—and to evaluate whether eGFR reporting in itself has any impact on patient care,” explained the study’s lead author, Brenda Hemmelgarn, MD, PhD, associate professor at the University of Calgary in Canada. “This could also be considered as a tool for other reporting systems and management of other kinds of clinical conditions.”
eGFR reporting was implemented voluntarily in outpatient settings throughout the province of Alberta as of October, 2004. The Alberta Kidney Disease Network, a collaborative network of nephrology researchers in Calgary and Edmonton, in which Hemmelgarn’s research group participates, sent physicians in the province an information sheet prior to implementation. The mailer described the rationale for eGFR reporting, gave basic recommendations for referral and management, and directed clinicians to a website for further information. For patients with eGFR ≥60 mL/min/1.73m2, eGFR was not reported, but a comment was included on lab results indicating that CKD is defined by persistent eGFR levels below that threshold. Conversely, for patients with eGFR <60 mL/min/1.73m2, labs not only reported eGFR but also included a comment that guidelines recommend that patients with eGFR <30 mL/min/1.73m2 be referred to a nephrologist.
For their study, Hemmelgarn and her colleagues used administrative and lab data on 1,135,968 adults in the province who had at least one outpatient serum creatinine measurement in the study period, which ranged from 2002-2007 or 2003-2007 depending on the region of Alberta where individuals lived. The researchers calculated eGFR for lab results from before Alberta’s implementation of eGFR reporting, and used lab-reported measurements after that time. The primary outcome was the monthly rate of first outpatient visits to a nephrologist. The researchers also looked at monthly rates of first outpatient visits to internists and general practitioners, and the association between eGFR reporting and hospitalizations for a primary discharge diagnosis of acute myocardial infarction, stroke, or congestive heart failure, as well as the monthly rates of new and any use of cholesterol-lowering medications. For patients at least 66 years old, the researchers also examined monthly rates of angiotensin-converting enzyme (ACE) inhibitor and angiotensin II receptor blocker (ARB) medications.
The investigators found that after eGFR reporting was implemented, the rate of first outpatient visits to nephrologists for patients with CKD increased significantly by 17.5 visits per 10,000 CKD patients per month—corresponding to a relative increase from baseline of 68.4%—and a slight but statistically significant decrease in the rate of new visits over time. No such associations were found among patients without CKD. The increased rate of first nephrologist visits after implementation of eGFR reporting was most pronounced among patients with eGFR <30 mL/min/1.73m2, predominantly in women, patients either 46-65 years old or ≥86 years old, and those with hypertension, diabetes, and other comorbidities. There was no association between eGFR reporting and increased rates of internal medicine visits or greater use of ACE inhibitors or ARB medications.
“The results are telling us that physicians must have actually been reading the prompts that were on the reports, taking a look at what the new eGFR was, and then referring for specialist care as indicated,” said Hemmelgarn.
But the results also point to flaws in eGFR-based CKD definitions and reporting, according to Richard Glassock, MD, professor emeritus of medicine at the David Geffen School of Medicine at University of California, Los Angeles. “The findings are not too surprising. They just document the frequency that the diagnosis of apparent chronic kidney disease is made when you embark on this recommended, and in some places, mandatory, calculation of estimated GFR and apply it to a classification criteria.” Glassock, who wrote the editorial accompanying Hemmelgarn’s study, contends that the KDOQI-recommended <60 mL/min/1.73m2 cutoff does not account for the natural decline in GFR that occurs with aging and is not necessarily a harbinger of progressive CKD. “It leads to a very high prevalence of the older population as having chronic kidney disease even though the data that that diagnosis will lead to benefits such as preventing end stage renal disease, death, or cardiovascular disease is extremely limited or non-existent,” he said.
Hemmelgarn agreed with Glassock that some revisions to eGFR thresholds are in order, but emphasized that the study in question was not designed to address that issue. However, her research team has looked at refining eGFR. “The big thing that’s come out of that is the importance of including a measure of proteinuria in the staging of chronic kidney disease,” she said. “It’s an important prognostic marker.” Both Glassock and Hemmelgarn indicated that KDOQI has assembled a working group to re-examine the 2002 schema and make recommendations concerning a possible revision.
Glassock contends that labs should not report eGFR until KDOQI revises its CKD definitions. “In my opinion, at the present time such reporting should not occur. Labs should report the serum creatinine value and reference ranges—preferably different ones for women and men—and let physicians decide if they are going to calculate eGFR or not, or what formula they’re going to use,” he said.
In contrast, Hemmelgarn believes reporting of eGFR, even with some shortcomings in the current KDOQI-recommended thresholds, has value. “I think it’s enhanced the recognition of chronic kidney disease. Obviously, we’ll need to revisit what’s being reported, but it has improved recognition and hopefully long-term outcomes, which we weren’t able to evaluate in the context of this setting. With a longer-term analysis we hope to see whether there’s an impact on morbidity and mortality in practice,” she said.
Regardless of the eGFR formulas or thresholds used, Hemmelgarn emphasized the importance of close ties among laboratorians, clinicians and researchers around the reporting of kidney function measurements. “It’s crucial to involve them all as a team,” she said. “Collaboration is critical, especially when you talk about measuring estimated GFR.”
Join us for a discussion of this article, and interact with your peers, on LinkedIn.