Skip To Main Content
 
 Table AICD-9-CM and ICD-10-CM codes for Chronic Kidney Disease (CKD) stages(introduced in 2006)
 Table 2.1Demographic characteristics of all patients, among Medicare (aged 65+ years) , Optum Clinformatics™ (all ages) and Veterans Affairs (all ages) patients, 2015
 Table 2.2Prevalence of comorbid conditions by diagnosis codes (CKD, CVD & DM), (a) total & (b) one or more, among Medicare (aged 65+ years) , Optum Clinformatics™ (aged 22-64 years) and Veterans Affairs (aged 22-64 years) patients, 2015
 Table 2.3Percent of patients with CKD by demographic characteristics, among individuals (aged 65+ years) in NHANES (2011-2015), Clinformatics (2015), Medicare 5% sample (2015) and Veterans Affairs (2015) datasets
 Table 2.4Prevalence of CKD, by demographic characteristics and comorbidities, among Medicare 5% sample (aged 65+ years), Optum Clinformatics™ (all ages) and Veterans Affairs (all ages) patients overall, and with DM or HTN, 2015
 Figure 2.1Prevalence of CKD among Medicare 5% sample (aged 65+ years) and Optum Clinformatics™ (all ages) patients, 2015
 Figure 2.2Trends in prevalence of recognized CKD, overall and by CKD stage, among Medicare patients (aged 65+ years), 2000-2015
 Table 2.5Change in CKD status from 2010 to 2015, among Medicare patients (aged 65+ years) alive and without ESRD in 2010
 Figure 2.3Trends in percent of patients with testing of urine albumin (a) in Medicare 5% sample (aged 65+ years) & (b) Optum Clinformatics™ (aged 22-64 years) patients without a diagnosis of CKD by year, 2005-2015
 Figure 2.4Trends in percent of patients with testing of urine albumin in (a) Medicare 5% (aged 65+ years) & (b) Optum Clinformatics™ (aged 22-64 years) patients with a diagnosis of CKD by year, 2005-2015
 Table 2.6Percent of patients with a physician visit in 2015 after a CKD diagnosis in 2014, among Medicare 5% patients (aged 65+ years)
 Figure 2.5Percent of CKD patients in 2014 with physician visit (nephrologist, PCP, both and neither), with lab testing in the following year (2015), by comorbidity
Download
Search This Page
Search All
Translate

Chapter 2: Identification and Care of Patients With CKD

  • Over half of patients in the Medicare 5% sample (aged 65 and older) had at least one of three diagnosed chronic conditions – chronic kidney disease (CKD), cardiovascular disease (CVD), or diabetes mellitus (DM), while 18.5% had two or more of these conditions. Within a younger population derived from the Optum Clinformatics™ Data Mart (ages 22-64 years), 9.9% had at least one of the three conditions, and 1.3% had two or more of these conditions. As indicated by diagnosis claims from the Department of Veterans Affairs (VA), 15.4% of patients had at least one of the three conditions, while 2.7% had at least two. (Table 2.2.b).
  • In the Medicare 5% sample and VA data, 11.7% and 9.7% of patients had a diagnosis of CKD in 2015, as opposed to only 1.1% of patients in the Optum Clinformatics™ population (Table 2.4).
  • The proportion of patients with recognized CKD in the Medicare 5% sample has grown steadily, from 2.7% in 2000 to 11.7% in 2015 (Figure 2.2).
  • Of those in the 2010 Medicare 5% sample who had a diagnosis of CKD Stage 3, by 2015 3.5% had progressed to end-stage renal disease (ESRD) and 40.3% had died. For these Medicare patients without identified CKD, progressions to ESRD and death by 2015 were 0.2% and 21.3% (Table 2.5).
  • Testing for urine albumin is recommended for patients with DM. Among Medicare patients with a diagnosis of DM, claims data indicated that testing for urine albumin has become more common, but was still conducted for less than half of these patients--40.5% in 2015, up from 24.8% in 2005. In 2015, urine albumin testing was performed in 48.6% of diabetic Medicare patients who also had diagnoses of CKD and hypertension (HTN). Patterns were similar in the Optum Clinformatics™ population, but with somewhat lower rates of testing (Figures 2.3 and 2.4).
  • Among Medicare patients with recognized CKD in 2014, patients who saw a nephrologist were roughly twice as likely to have a claim for urine albumin testing in 2015 (53.1%) than those who saw only a primary care physician (25.4%; Figure 2.5).

Introduction

Epidemiological evaluations of the identification and care of patients with chronic kidney disease (CKD) are a significant challenge, as unlike ESRD, no single data source contains all the information necessary to definitively identify CKD-related care practices in the United States (U.S.) population. Furthermore, most large administrative health care datasets lack the biochemical data (serum creatinine and urine albumin or urine total protein) required per KDIGO guidelines for definitive identification of CKD.

As presented in Volume 1, Chapter 1, CKD in the General Population, The National Health and Nutrition Examination Survey (NHANES) is a nationally representative survey that contains the biochemical information with which to estimate the prevalence of CKD in the U.S. However, NHANES is constrained by its cross-sectional nature, a relatively small sample size, and lack of geographic detail. This limits precision in estimating prevalence, evaluating long-term outcomes, adverse events, and quality of care delivered, and the ability to conduct analyses by geography or on subsets of patients.

In addition, the NHANES includes only a single measure of serum creatinine and urine albumin for each patient. Per KDIGO guidelines, two abnormal measures over at least 90 days are necessary to definitively determine CKD. Because NHANES-based calculations rely on laboratory measures at a single time point, they may overestimate the national prevalence of CKD. Regardless, NHANES is generally considered the best available source of such information at the present time.

To provide a more comprehensive picture of the identification and care of CKD throughout the nation, in this chapter we compliment NHANES with the examination of health care data in large and diverse administrative health care datasets—the Medicare 5% sample, and data from the Optum ClinformaticsTM Data Mart and from the U.S. Department of Veterans Affairs Health System (VA).

We first present the prevalence of CKD in these health system populations as recognized through diagnosis claims—both for the overall disease state and with the comorbidities of DM and HTN. This was achieved through comparison of rates in the NHANES, Medicare 5% sample, Optum Clinformatics™ and VA populations among cohorts of patients aged 22-64, or 65 and older. These were stratified by demographic characteristics in order to highlight issues with identification of CKD across these various types of data.

We next examined longitudinal changes in CKD status and general outcomes for patients at high risk for kidney disease, through presenting trends in laboratory screening and monitoring of patients with and without CKD. Finally, we assessed the spectrum and impact of follow-up care received by newly diagnosed CKD patients.

Methods

For this year’s chapter we utilized several large health care datasets. The general Medicare 5% sample includes an average of 1.2 million patients each year. The Optum Clinformatics™ Data Mart cohort was drawn from the commercial plans of a large U.S. national health insurance company, and holds information on about nine million lives per year. The national health system-derived data from the U.S. Department of Veterans Affairs (VA) represents more than six million veterans.

Analyses using the Medicare 5% dataset are restricted to patients aged 65 and older and are limited to those persons with both Part A and Part B fee-for-service coverage. Persons covered by Medicare managed care programs are not included in this source because of the absence of billing claims. The Optum Clinformatics™ Data Mart data provides insight into a younger, employed population and their dependent children. Like Medicare data, it contains diagnosis and procedure codes as found on claims. The Optum Clinformatics™ dataset also includes information on pediatric age groups, although for some analyses in this chapter only adult patients (ages 22-64 years) are included. Finally, the VA dataset includes both diagnosis and procedure codes and more complete biochemical test data. This allows us to estimate the prevalence of CKD as indicated by diagnoses codes combined with serum creatinine blood and urine test results wherever available.

Throughout this chapter, the term ‘recognized CKD’ is used when patients are identified based on the presence of a relevant diagnosis code in Medicare, Optum Clinformatics™, or VA data. This implies that either a provider or billing coder in the health care system recognized the presence of CKD. As such, prevalence of ‘recognized CKD’ likely underestimates true disease prevalence. An observed trend may not necessarily indicate a true change in disease prevalence, but rather a change in clinical awareness or recognition of CKD, or indeed, evolving billing practice. Studies have shown that diagnosis codes for CKD generally have excellent specificity (>90%), though their sensitivity is low (Grams et al., 2011).

To identify the recognized CKD population we included a variety of ICD-9-CM diagnosis codes, some of which are sub-codes under related comorbidities such as DM (250.4x) and HTN (403.9x), and other conditions that are kidney-disease specific, such as glomerular disease (583.x). In 2006, new CKD stage-specific codes (585.x) were introduced, providing an opportunity to track trends in the severity of CKD over time. Since their introduction, the CKD stage-specific codes have been increasingly utilized, accounting for 49% of all CKD diagnostic documentation in 2007 and 68% in 2015.

Beginning on October 1, 2015, the new ICD-10-CM coding system was implemented, and its related diagnosis codes were then utilized to identify CKD stages and comorbid conditions. Table A lists the CKD-related ICD-9-CM and ICD-10-CM codes used in this chapter.

Details of this data are described in the Data Sources section of the CKD Analytical Methods chapter.

See the CKD Analytical Methods section of the CKD Analytical Methods chapter for an explanation of the analytical methods used to generate the study cohorts, figures, and tables in this chapter. Microsoft Excel and PowerPoint files containing the data and graphics for these figures and tables are available to download from the USRDS website.

Table A ICD-9-CM and ICD-10-CM codes for Chronic Kidney Disease (CKD) stages(introduced in 2006)

Patients Characteristics across Datasets

Table 2.1 presents demographic and comorbidity characteristics of individuals in the Medicare 5% sample (aged 65 and older), the Optum Clinformatics™ dataset (all ages) and in data from the VA health system. The mean age of Medicare patients was 74.6 years, the mean age of Optum Clinformatics™ patients was 35.6 years, and for U.S. Veterans was 62.4 years. The high prevalence of comorbid conditions in the Medicare 5% sample reflects the older age of these patients. For example, 59% and 24% of the Medicare sample have diagnoses of HTN or DM. In comparison, only 10.3% and 4.4% of the total Optum Clinformatics™ population have diagnoses of HTN or DM. In VA data these proportions were 25.5% (HTN) and 16.9% (DM).

Table 2.1 Demographic characteristics of all patients, among Medicare (aged 65+ years) , Optum Clinformatics™ (all ages) and Veterans Affairs (all ages) patients, 2015

Table 2.2 provides the prevalence of recognized CKD, DM, and cardiovascular comorbid conditions among patients aged 65 and older in the Medicare population, for Optum Clinformatics™ adults aged 22 through 64 years, and VA patients (ages 22 to 64). Younger Optum Clinformatics™ patients were excluded as these comorbidities are rare in this population. Of Medicare patients aged 65 and older, recognized (i.e., coded diagnosis of) CKD was observed in 11.7%. Over half of the Medicare cohort (51.2%) had at least one of these comorbid conditions, 18.5% had two or more, and 4.1% had all three. As expected, the prevalence of recognized CKD in the Optum Clinformatics™ population was substantially lower, driven by the lower prevalence among younger patients. Approximately 9.9% of this cohort had at least one of these comorbid conditions, and 1.3% had two or more.

Table 2.2 Prevalence of comorbid conditions by diagnosis codes (CKD, CVD & DM), (a) total & (b) one or more, among Medicare (aged 65+ years) , Optum Clinformatics™ (aged 22-64 years) and Veterans Affairs (aged 22-64 years) patients, 2015

Comparison of CKD Prevalence across Datasets

Table 2.3 compares the prevalence of CKD in the NHANES, Medicare 5% sample, Optum Clinformatics™ and VA populations among patients aged 65 and older. We stratified by demographic characteristics in order to highlight issues with identification of CKD in the varying types of data. Across all datasets, the prevalence of CKD increased with older age. Variance between the data sources, however, can somewhat be explained by the nature of their measurements and specific populations.

The absolute prevalence of CKD was highest in the NHANES data, intermediate in the VA data (eGFR-based), and lowest when based on diagnosis codes alone in Medicare claims, Optum Clinformatics™, or VA data.

The NHANES, by design, includes laboratory measurement of kidney function in all participants, thus providing the closest estimate of the true prevalence of CKD. Overestimation is possible, however, because it relies on a single measurement. NHANES also does not represent people living in long-term care facilities—many of those residents have Medicare insurance and were represented in the Medicare 5% sample.

The prevalence of recognized CKD based on diagnosis codes is lowest due to under-recognition and likely under-coding of the condition, particularly in its earlier stages, with more accurate capture of advanced cases of CKD.

For the VA population, CKD prevalence is presented based on diagnosis codes and available laboratory data documenting at least one serum creatinine result corresponding to an eGFR <60 ml/min/1.73m2. Blood and urine assays are initiated by clinical indication and not performed in all patients, and thus likely underestimate the true prevalence in the population served by the VA health system.

The overall CKD prevalence, and CKD prevalence by gender and race/ethnicity varies substantially depending on the method of CKD ascertainment: survey (NHANES), vs. claim-based (Medicare and Optum ClinformaticsTM), vs. claim and lab based data (VA data).

Table 2.3 Percent of patients with CKD by demographic characteristics, among individuals (aged 65+ years) in NHANES (2011-2015), Clinformatics (2015), Medicare 5% sample (2015) and Veterans Affairs (2015) datasets

Table 2.4 presents the prevalence of recognized CKD by demographic characteristics and comorbidities in the Medicare, Optum Clinformatics™ and the VA populations, overall and with DM or HTN. The prevalence of recognized CKD increased with age in all three datasets, and from 8% at ages 65–74 to 20.3% at age 85 and older in the Medicare data. Males had slightly higher prevalence than females in the Medicare and Optum ClinformaticsTM datasets, but there was substantially higher prevalence in women than men in the VA dataset.

The prevalence of CKD among Blacks/African Americans was higher than Whites in the Medicare and Optum ClinformaticsTM datasets, but lower in the VA dataset. Results from adjusted analyses of the Medicare dataset (data not shown) confirm greater odds of recognized CKD in older patients, Blacks, and those with DM, HTN, or cardiovascular disease. Among Optum ClinformaticsTM patients of age comparable to the Medicare population, the prevalence remained lower, possibly reflecting a healthier, employed population. As expected, the prevalence of recognized CKD was higher in both datasets among those with a diagnosis of DM or HTN, and particularly so in the younger patients in the Optum Clinformatics™ dataset.

Table 2.4 Prevalence of CKD, by demographic characteristics and comorbidities, among Medicare 5% sample (aged 65+ years), Optum Clinformatics™ (all ages) and Veterans Affairs (all ages) patients overall, and with DM or HTN, 2015

Figure 2.1 presents maps displaying the prevalence of recognized CKD by state, in the Medicare 5% sample and the Optum Clinformatics™ dataset. Variation in prevalence across states was more than two-fold in both datasets.

Figure 2.1 Prevalence of CKD among Medicare 5% sample (aged 65+ years) and Optum Clinformatics™ (all ages) patients, 2015

Figure 2.2 shows the 2000-2015 trend Medicare in prevalence of recognized CKD overall and by CKD stage-specific code. The prevalence of recognized CKD has steadily risen each year, accompanied by a comparable increase in the percentage of patients with a stage-specific CKD diagnosis code.

Figure 2.2 Trends in prevalence of recognized CKD, overall and by CKD stage, among Medicare patients (aged 65+ years), 2000-2015

Longitudinal Change in CKD Status and Outcomes, Based on Diagnosis Codes

Table 2.5 shows patient status of CKD stage, ESRD, or death in 2014-2015 for those who had a CKD diagnosis in 2010. Among patients with no CKD in 2010, 21.4% had died after five years, while 0.2% had reached ESRD and 0.1% were alive with ESRD by the end of 2015. In comparison, patients with a CKD diagnosis in 2010 were much more likely to have these outcomes. Among CKD patients, by 2015 43% had died, 4% had reached ESRD, and 1.8% were alive with ESRD.

Table 2.5 Change in CKD status from 2010 to 2015, among Medicare patients (aged 65+ years) alive and without ESRD in 2010

Laboratory Testing of Patients with and Without CKD

Assessing the care of patients at high risk for kidney disease has long been a focus of the USRDS, and is part of the Healthy People 2020 goals developed by the Department of Health and Human Services (see the Healthy People 2020 chapter). There are no recommendations to screen asymptomatic patients, but individuals at high risk for CKD, most notably those with DM, should be screened periodically for kidney disease; those with CKD should be monitored for progression of disease.

Urine albumin is a valuable laboratory marker used to detect signs of kidney damage and to evaluate for disease progression. Serum creatinine measurement is usually included as part of a standard panel of blood tests, but urine albumin testing must be ordered separately. For this reason urine albumin testing may better represent intent to assess kidney disease.

The American Diabetes Association (ADA) recommends urine testing for albumin in patients with DM. The 2012 Kidney Disease Improving Global Outcomes (KDIGO) guidelines on CKD evaluation and management recommend risk stratification of CKD patients using both the urine albumin/creatinine ratio and the estimated eGFR (based on estimating equations incorporating serum creatinine values). They emphasized that these tests are needed to understand patients’ kidney disease status and risks of death and progression to ESRD (Matsushita et al., 2010; KDIGO CKD Work Group, 2012).

As shown in Figure 2.3, 12.3% of Medicare patients without diagnosed CKD received urine albumin testing in 2015, while 3.6% of Optum Clinformatics™ patients aged 22 to 64 years without diagnosed CKD received a urine albumin test (assessment of urine protein was also included in these percentages, representing approximately 20% of testing performed). Among Medicare patients, 40.5% with DM alone had urine albumin testing, compared to 6.3% of patients with HTN alone.

Having both DM and HTN is known to increase the likelihood of developing CKD. Among Medicare beneficiaries without a CKD diagnosis, 41.5% had urine albumin testing in 2015. Similar patterns were seen in the Optum Clinformatics™ population—37.7% of patients with DM alone in 2015 had urine albumin testing, compared to 5.6% with HTN alone, and 38.9% with both DM and HTN.

Figure 2.3 Trends in percent of patients with testing of urine albumin (a) in Medicare 5% sample (aged 65+ years) & (b) Optum Clinformatics™ (aged 22-64 years) patients without a diagnosis of CKD by year, 2005-2015

As shown in Figure 2.4, patients with a diagnosis of CKD received testing at similar, though somewhat higher rates, to patients without CKD. In 2015, among patients with the combined diagnoses of CKD, DM, and HTN, urine albumin testing was performed for 48.6% of the Medicare and 47% of the Optum Clinformatics™ cohorts.

Figure 2.4 Trends in percent of patients with testing of urine albumin in (a) Medicare 5% (aged 65+ years) & (b) Optum Clinformatics™ (aged 22-64 years) patients with a diagnosis of CKD by year, 2005-2015

Physician Visits After a CKD Diagnosis

Table 2.6 indicates the percentage of patients with a CKD diagnosis in 2014 who had at least one visit to a primary care physician, cardiologist, or nephrologist in 2015. Patients with any CKD diagnosis were far more likely to visit a primary care physician or a cardiologist than a nephrologist. This may relate to the fact that most guidelines, including KDIGO CKD, indicate the need for referral to nephrology only for those with advanced, Stage 4 CKD (see Table A), unless there are other concerns such as rapid progression of disease. Indeed, fewer than one-third of patients with any CKD claim in 2014 were seen by a nephrologist in the subsequent year. However, nearly half with CKD Stage 3 and roughly two-thirds with CKD Stage 4 or higher visited a nephrologist in 2015. Whether the involvement of a nephrologist improves outcomes, and at what stage of CKD, is a matter of ongoing research interest.

Overall, the patterns of physician visits varied little across demographic categories. A notable exception was that patients aged 85 and older with CKD Stage 3 or higher were as likely as younger patients to visit a cardiologist, but less likely to visit a nephrologist.

Table 2.6 Percent of patients with a physician visit in 2015 after a CKD diagnosis in 2014, among Medicare 5% patients (aged 65+ years)

Figure 2.5 illustrates the proportion of patients with CKD in 2014 who were tested for urine albumin in 2015, according to whether they saw a primary care physician or nephrologist in 2014. Patients who saw a nephrologist were more likely to be tested for urine albumin than those who saw only a primary care physician. This difference was greatest for those without DM. Diabetic patients showed a smaller difference in testing for urine albumin across provider type. This finding relates to the wide dissemination of guidelines for routine renal function assessment in diabetics that are directed at primary care physicians by organizations such as the American Diabetes Association.

Figure 2.5 Percent of CKD patients in 2014 with physician visit (nephrologist, PCP, both and neither), with lab testing in the following year (2015), by comorbidity

References

Grams ME, Plantinga LC, Hedgeman E, et al. Validation of CKD and related conditions in existing data sets: a systematic review. Am J Kidney Dis 2011;57:44-54.

Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013;3(1):1–150.

Matsushita K, van der Velde M, Astor BC, et all. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 2010;375:2073–2081.