Abstract
Introduction
The risk of Chronic Kidney Disease (CKD) increases as we age, with 1 in 2 people over the age of 75 having CKD. Previous research in London found 48% of those with CKD stage 3-5 were undiagnosed. The commonest cause of death in patients with CKD is cardiovascular events. This means that timely detection of CKD is important to allow risk-stratification and monitoring of other cardiovascular risk factors, which could help improve mortality of those with CKD.
This audit consists of two cycles to see whether those 75 years or older who fit the NICE CKS criteria for CKD are diagnosed.
Methods
In the first cycle, patients aged 75 or older who had an eGFR between 30-60 on a single blood test in the first two weeks of April 2025 were identified and analysed. The results and how to diagnose CKD were then presented to clinical staff at the GP practice. A poster was created to help staff interpret eGFR results and posted in clinical areas. Then a second cycle was conducted over the first two weeks of July 2025.
Results
In the first cycle, 79 patients were identified. Of those 79 patients, 69 fit the CKS criteria for CKD. However, 29% (20 patients) had no previous diagnosis of CKD. In the second cycle, 83 patients were identified. Of those 83 patients, 68 patients fit the CKS criteria for CKD; however 35% (24 patients) had no previous CKD diagnosis.
Conclusions and next steps
From these two cycles, it is clear that CKD is underdiagnosed in primary care despite interventions to increase awareness. The next steps are to re-audit again to see if the interventions needed more time before having an impact, and do smaller workshops to improve understanding and detection of CKD at the GP practice.
Comments
An interesting piece of work
Thank you for taking the time to do work in this area. Currently I am working in the community in a Frailty Hospital at Home team. For our notes, we use a combination of hospital-based IT systems, and community-based IT systems, specifically SystemOne and EMIS.
Did you explore whether the issue was clinicians not recognising CKD, or recognising it but not coding it? From my experience at least in primary care, missed or incorrect coding can hugely affect prevalence figures in data capute
I think it was a combination…
I think it was a combination of both! I believe clinicians checking bloods would assume older patients had CKD, but didn't document/code or do follow up investigations/management (i.e. urine ACR, checking hypertension or diabetes). I looked at system one notes to see whether there were any documentations regarding CKD in patients with undiagnosed CKD and majority of them had no mention of kidney disease in their notes! When I checked what bloods with reduced eGFR were filed as most of the clinicians filed these bloods as 'abnormal but as expected' - which suggests they were aware the patient had CKD but didn't investigate further. Also one clinician may review the first set of bloods, but the repeat bloods may not be reviewed by the same clinician.
But I also agree that coding was also an issue. I found several different coding options for CKD existed (renal failure, kidney failure, all stages of CKD) which probably makes it harder to code.
While I was at the GP practice I also found coding tricky, sometimes patients were miscoded and I found getting that coding changed quite difficult.