Dr Shweta Bansal Highlights Lack of Screening, Diagnoses of CKD

Many of the patients in our clinic system were not being screened for chronic kidney disease even though they exhibited risk factors, said Shweta Bansal, MD, FASN, associate professor of medicine in the nephrology division at the University of Texas Health Center at San Antonio.

Transcript:

Why did you chose to research factors associated with chronic kidney disease (CDK) screening?

This project basically, the background for this project, was that I strongly felt that in our clinic system when we get referrals for early CKD, or patients with hypertension who do not have controlled hypertension, that they were not on diuretics. As you know, that salt retention or hypovolemia plays a huge role in the pathogenesis of hypertension in the general population and definitely in CKD.

If volume retention is a means of pathogenesis, then either a low salt diet or diuretics will be the treatment. And you know how it isn't easy to achieve low salt diet when every single thing you buy from outside has salt already, whether cooked or uncooked. So that's not very practical if you do not cook from scratch every day, 3 times a day. You end up having that high salt, so the role of diuretics becomes very important. When we are seeing these patients in our clinic, many of those patients are not on diuretics. I just wanted to see, what is the reason that these patients are not on diuretics? What is the percentage of the use and what are the reasons behind it? That was a whole background, I started working on it.

First, I had to look at how many CKD patients are out there with uncontrolled hypertension and how many of them are on diuretics. During that project, I realized that many of these patients, they're not even getting screened for CKD. And then many of them, then if they were screened, they will not have the diagnosis of CKD in their chart. So the question was, they just had that GFR test done for some other reason, or in primary care were they looking for CKD screening. That's how this whole thing started. Because my main thing was, how many patients with these uncontrolled hypertension, with CKD are on diuretics. But then I realized many of these patients, uncontrolled hypertension, they are not even being looked for CKD, or if they were looked for they were not even recognized for CKD. So that's how this whole project started.

How did you carry out your study?

I was working at Veterans Affairs (VA) mostly at that time. The second thing, VA has this very defined data warehouse, so I can easily access a big chunk of the population. On the other hand, for other hospital care systems or medical systems, you may not have a very defined cohort. That's why I chose to do VA. CKD screening has been recommended in patients who are at high risk of developing CKD, so that includes basically mostly patients with diabetes, hypertension, with previous family history or heart disease and then ethnic minority races. I chose to stay just with patients with diabetes and hypertension, because these are responsible for 80% and 90% of the CKD instances. So I just focused on these 2 groups, because they have very defined ICD 9 and 10 criteria. I can have a very clean database. I chose to have patients who have been seen in primary care clinics, and to ensure that these are continuity patients for that primary care clinic. They just did not come one time for some pro re nata (PRN) issues. We chose those patients who were seen in primary care clinics at least twice within 18 months. These are the kind of continuity of primary care patients then who have diabetes and hypertension. Our duration of cohort was basically October 2012 to September 2019.

Once we had that cohort, I think the final cohort number was about 270,000 patients, we looked at charts for documentation of serum creatinine or estimated glomerular filtration rate (EGFR), which kind of tells you about the CKD and presence of urine protein or albumin along with urine creatine. We looked at those values in the chart how many times patients had those documented. Out of those patients who have these documented, then we looked at how many had EGFR below 60 and how many had urine albumin creatinine ratio more than 30 milligram per gram, which these 2 criteria define CKD. To call them CKD, you have to have these values at least twice, 90 days apart. That's how we kind of first identified our study cohort.

Then we looked at the CKD screening how many of these patients had the EGFR or urine albumin in the chart, and then we looked at how many had defined CKD. In those patients who had these defined CKD by those definitions, we looked at how many of them had the ICD 9 or 10 codes for the CKD. It could be generalized CKD or hypertensive CKD or diabetic CKD. There are a variety of codes. We looked for those as well as if they had a reference for nephrologist. That kind of tells us that those primary care providers thought about, they looked at those screening tests, they identified that patients have CKD, so they know that these patients have CKD. After that, once patients like these are screened versus not screened, these were recognized patients for CKD versus not recognized, then we look at a multitude of clinical variables, their demographics, their smoking history, or their race, or the time of entry into the cohort, or their CKD staging, or comorbidities. Then we also looked at how many times they were seen by the primary care or by specialty care or they had the procedure done, where they needed contrast, because then you need serum creatinine measures before contrast, or what medications they had. We kind of thought of all the possible factors which would be prompting some providers to order serum creatinine or EGFR to look for those factors and take those out. Then you're like, Okay, this test was ordered for the screening purposes rather than just as a bystander for something like that.