Is Awareness of Chronic Kidney Disease Associated with Evidence-Based Guideline-Concordant Outcomes?Tuot D.S.a, b · Plantinga L.C.d · Hsu C.a, b · Powe N.R.b, c
aDivision of Nephrology, and bDepartment of Medicine, University of California, and cCenter for Vulnerable Populations, San Francisco General Hospital, San Francisco, Calif., and dDepartment of Epidemiology, Emory University, Atlanta, Ga., USA
Background: Awareness of chronic kidney disease (CKD) is low. Efforts are underway to increase recognition of CKD among patients, assuming that such an increase will lead to better outcomes through greater adherence to proven therapies. Few studies have tested this assumption. Methods: CKD awareness, defined by a ‘yes’ answer to ‘Have you ever been told by a healthcare provider you have weak or failing kidneys?’, was assessed among 2,404 adults with CKD stages 1–4, who participated in the 2003–2008 National Health and Nutrition Examination Surveys. Odds of blood pressure (BP) control, self-reported use of angiotensin-converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB), and glycemic control, were determined among those aware vs. unaware of their CKD. Results: Optimal BP control, ACEI/ARB use and glycemic control were low in the US adult population with CKD, although there was a recent increase in attainment of guideline-concordant BP control. Odds of BP control and ACEI/ARB use were not different among individuals aware of their CKD compared to those unaware (adjusted odds ratio (AOR) 0.91; 95% CI 0.52–1.58 and AOR 0.75; 0.44–1.30, respectively). CKD awareness among diabetic participants was not associated with glycemic control (AOR 0.41; 95% CI 0.14–1.18). Conclusion: Awareness of CKD is not associated with more optimal BP control, ACEI/ARB use or glycemic control. Future efforts in this area should further explore the measurement of CKD awareness and behaviors associated with CKD awareness.
Copyright © 2012 S. Karger AG, Basel
Chronic kidney disease (CKD) is common  and associated with excessive mortality and morbidity. Better CKD management – through evidence-based medical therapies – improves outcomes . It is generally accepted that awareness of CKD by providers and patients is critical for implementation and adherence to those therapies.
Blood pressure (BP) control, use of an angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB), and glycemic control are three evidence-based CKD practice guidelines known to slow progression of CKD to end-stage renal disease (ESRD) and to reduce cardiovascular-related complications . National efforts to increase recognition of CKD among providers and patients  have been based partly on the premise that these efforts will lead to wider adoption of these therapies and, consequently, better health outcomes in patients with CKD.
Recent cohort studies have provided conflicting evidence regarding the association of primary care provider awareness of CKD with improved overall quality of CKD care, including greater prescription of ACEI/ARB and improved BP control [4,5]. Patient awareness of CKD, arguably a determinant of medication adherence and BP control, was not examined in either study and may provide one explanation for the contradictory results. While individual awareness of CKD is integrally related to provider awareness, the two entities are separate. The influence of patient awareness of CKD on outcomes concordant with CKD practice guidelines is therefore important to understand and remains a gap in the literature.
Our primary objective was to test the hypothesis that individual awareness of CKD was associated with improved BP control, ACEI/ARB use, and glycemic control in a nationally representative sample of US adults with CKD with recent contact with the healthcare system. Because previous studies have demonstrated low levels of guideline-concordant CKD care in the United States [6,7], we also hypothesized that prevalence of these outcomes would increase over time.
The National Health And Nutrition Examination Survey (NHANES) is conducted by the National Center for Health Statistics to examine trends in disease prevalence in cross-sectional representative samples of noninstitutionalized US civilian residents . We examined data from the 2003–2008 NHANES, after the National Kidney Foundation Disease Outcomes Quality Initiative (KDOQI) was introduced in 2002. All participants gave informed consent. The protocol was approved by an institutional review board.
The total number of adult nonpregnant 2003–2008 NHANES study participants with CKD stages 1–4 was 2,705. Individuals with an estimated glomerular filtration rate (eGFR) <15 ml/min/1.73 m2 were excluded from the outset, since guideline-concordant care would differ in this population. Our main study population also excluded individuals who did not self-report awareness or unawareness of CKD (n = 13) and did not report a routine site for health care or having seen a health care provider within the previous year (n = 288). This restriction allowed the study population (final n = 2,404) to represent a group of individuals receiving regular medical care, increasing the likelihood that awareness would be associated with guideline-concordant care.
The predictor variable was awareness of CKD, defined by a ‘yes’ response to the questionnaire item: ‘Have you ever been told by a doctor or other health professional that you have weak or failing kidneys (excluding kidney stones, bladder infections, or incontinence)?’ .
Outcomes included achievement of three evidence-based therapies known to help slow the rate of CKD decline: BP control, use of ACEI/ARB, and glycemic control (among individuals with DM). BP was considered controlled if the average measurement of systolic or diastolic BP was ≤130 or ≤80 mm Hg.  Glycemic control was defined by glycosylated hemoglobin <7.0%  among patients with self-reported DM.
CKD was defined using single assessments of eGFR and urinary albumin:creatinine ratios, according to the KDOQI staging guidelines . Estimated GFR was calculated according to the 4-variable Modification of Diet in Renal Disease (MDRD) Study equation for calibrated serum creatinine level .
CKD awareness (via questionnaire) and the number and type of anti-hypertensive medications (from prescription bottles provided by the participant) were obtained during NHANES interviews. BP was measured during the exam; the mean of all measurements (at least three) was used. Serum glycosylated hemoglobin was measured by high-performance liquid chromatography. Serum creatinine was measured by the modified kinetic method of Jaffe; levels were corrected for different analyzers [13,14]. Urine albumin was measured using a solid-phase fluorescence immunoassay; urine creatinine was measured using the modified Jaffe kinetic method . Self-reported sociodemographic variables and diagnoses were obtained during NHANES interviews.
Participant characteristics were compared by awareness and unawareness of CKD by χ2 and Wilcoxon rank-sum tests. Prevalences of outcomes concordant with evidence-based practice guidelines were estimated overall and by NHANES survey year. Variance of proportions was estimated with Taylor series linearization.
Multivariable logistic regression was used to estimate the independent association between CKD awareness and BP control, ACEI/ARB use, and glycemic control (among the subset with DM). Because severity of renal dysfunction is associated with greater CKD awareness  and potentially with greater difficulty in achieving outcomes consistent with guideline-concordant care, we included propensity scores (in quintiles) for CKD severity, determined by characteristics that were shown or thought a priori to be associated with awareness and severity of CKD. Each model was further adjusted for potential confounders that remained significantly different among the aware and unaware study participants within each quintile of propensity score. Sensitivity analyses with different study populations were performed to assess robustness of the results: (i) restricted to individuals with an eGFR <60 ml/min/1.73 m2; (ii) included all individuals with CKD irrespective of whether they had seen a healthcare provider in the previous year; and (iii) included individuals with CKD as defined by the CKD-EPI equation . Additionally, given the current debate about target BP control in individuals with CKD, a sensitivity analysis exploring a target BP <140 or <90 mm Hg among individuals without albuminuria was performed.
To obtain nationally representative population estimates, all analyses were performed with appropriate weighting using the ‘SVY’ commands in Stata version 11 (StataCorp, College Station, Texas, USA).
Only 7.4% of those with CKD and recent medical care self-reported awareness of CKD. Those who were aware were more likely to be non-Hispanic black and less likely to have completed a high school education. In addition to having more severe kidney dysfunction, individuals who were aware of their CKD were much more likely to have hypertension and self-reported diabetes (table 1).
|Table 1. Characteristics of the US population with CKD, stratified by awareness of CKD (NHANES 2003–2008)|
Only 39% achieved BP control and a mere 21% self-reported use of an ACEI/ARB. Among individuals with CKD and diabetes, 49% achieved glycemic control. Among the individuals with an eGFR between 15 and 59 ml/min/1.73 m2, 37% achieved BP control, 23% reported use of an ACEI/ARB, and, among those with diabetes, 57% achieved glycemic control.
There was a significant increase in achievement of BP control over time and a small but statistically significant decrease in ACEI/ARB use (fig. 1). Prevalence of glycemic control, however, did not change appreciably between 2003 and 2008. Awareness of CKD remained steady throughout this time period (plinear trend = 0.89).
|Fig. 1. US population prevalences of guideline-concordant CKD outcomes and temporal trend (NHANES 2003–2008).|
Overall, awareness of CKD was not associated with greater odds of achieving of any of the three evidence-based CKD therapies examined in this study. Awareness of CKD was not associated with increased likelihood of BP control (table 2), an association that was not modified by CKD stage (pinteraction = 0.18) or presence of albuminuria (pinteraction = 0.19). Odds of ACEI/ARB use and glycemic control were not different among those aware vs. unaware of their CKD, either overall or when stratified by eGFR or presence of albuminuria (table 2).
|Table 2. Odds ratios (OR) of outcomes consistent with guideline-concordant care, if aware vs. unaware of chronic kidney disease|
Comparable results were noted when the study population was restricted to individuals with eGFR <60 ml/min/1.73 m2 (n = 1,410). Awareness of CKD was not associated with increased likelihood of BP control (adjusted odds ratio (AOR) 0.80; 95% CI 0.43–1.48), ACEI/ARB use (AOR 1.0; 95% CI 0.99–1.02), or glycemic control among those with diabetes (AOR 0.34; 95% CI 0.09–1.32). Similarly, when the study population was expanded to include individuals with CKD who had had not seen a healthcare provider within the previous year or did not report a routine site for health care, awareness of CKD was not associated with greater achievement of guideline-concordant outcomes. Again, these associations were not modified by CKD stage or presence of albuminuria (data not shown). Analyses repeated with the CKD-EPI equation to define CKD also showed similar results (table 2).
When target BP was changed to ≤140 or ≤90 mm Hg among individuals without albuminuria, overall prevalence of BP control was higher in all years (40.2% in 2003–2004, 48.3% in 2005–2006, 60.4% in 2007–2008). Again, there was a statistically significant increase in achievement of BP control over those years (plinear trend < 0.001). However, odds of BP control was not associated with awareness of CKD (AOR 0.84; 95% CI 0.50–1.42).
There are two main findings in this study. First, we found that – contrary to our hypothesis – individual awareness of CKD was not associated with better BP or glycemic control or more use of ACEI/ARB. Second, although we newly demonstrate an increase in attainment of BP control over time, we document overall low prevalences of BP control, ACEI/ARB use and glycemic control in the US adult population with CKD using recent NHANES data. This adds to prior literature documenting limited achievement of guideline-concordant care among study populations [7,16,17].
Individual awareness of chronic diseases, such as diabetes, hypertension, and congestive heart failure, has been associated with greater individual engagement in health care and adherence to medical therapies [18,19,20,21,22]. Educational interventions among patients nearing renal replacement therapy have also been associated with improvements in health outcomes, such as delays in initiation of dialysis and increased overall survival [23,24]. Thus, there is strong reason to believe that individual awareness of CKD would be associated with improved intermediate outcomes, such as achievement of guideline-concordant CKD care. Our findings that participant awareness of CKD was not associated with greater achievement of any of the three outcomes examined, were unexpected.
These findings, consistent with data from the Kidney Early Evaluation Program (KEEP) that have also demonstrated lack of association between CKD awareness and odds of BP control , may be explained in a variety of ways. Of course, it is possible that individual awareness of CKD does not actually influence achievement of guideline-concordant CKD care, contrary to what has been seen in other diseases. However, our data may also result from factors that obscure the true association between awareness and attainment of guideline-concordant care. Perhaps provider recognition of CKD and subsequent prescription of appropriate evidence-based CKD therapies play a far greater role than individual awareness of CKD in assuring good BP control, ACEI/ARB use and glycemic control. If individual awareness of CKD is not associated with provider prescription of CKD therapies, its influence on the outcomes studied would be minimal, consistent with our results. This could occur if health care providers are not fully aware of the CKD guidelines, do not believe the evidence behind the recommendation of such therapies, or believe that the prescription of such therapies is futile. It is also conceivable that providers become aware of, or focus on, their patients’ CKD only at very advanced stages, at which point achievement of good BP and glycemic control and ACEI/ARB therapy may be limited by side effects related to low eGFRs.
Additionally, our results may be due to poor discrimination among individuals who are aware vs. unaware of their CKD. Ascertainment of CKD awareness in the NHANES survey is through the following question: ‘Have you ever been told by a doctor or other health professional that you have weak or failing kidneys (excluding kidney stones, bladder infections, or incontinence)?’ While this question may be very specific for CKD awareness – only 1.5% of individuals without CKD responded ‘yes’ to being aware of having CKD – it may lack sensitivity. Inclusion of individuals in the ‘unaware’ group, rather than the ‘aware’ group would result in misclassification and bias our study results towards the null.
The NHANES question is the only metric currently used to estimate population trends in CKD awareness. Our results raise concern that the question may have poor performance characteristics, i.e., it may be very insensitive, particularly among individuals with limited health literacy, who have lower objective knowledge about kidney disease . To the best of our knowledge, the question used in NHANES has not been validated against a gold standard. Further, it lacks face validity. In an informal survey of practicing nephrologists, none volunteered that they use the terms ‘weak or failing’ kidneys to describe chronic kidney disease, except when patients are very close to needing dialysis (i.e. stage 5 CKD). While our study outcomes – achievement of guideline-concordant CKD care – are imperfect benchmarks against which this question should be evaluated, we believe that they offer insight into the limitations of this question.
There are several limitations to this study. Causality cannot be inferred because of the cross-sectional nature of NHANES data. Also, reverse causality (e.g., difficult-to-control BP leading to increased awareness of underlying CKD among providers and patients) cannot be ruled out. Further, the association between individual awareness of CKD and achievement of evidence-based medical therapies may be confounded by severity of CKD, as it is much more difficult to achieve BP control, maintain daily ACEI/ARB use, and achieve glycemic control in individuals with low eGFRs. While we used propensity scores to account for CKD severity, residual confounding may still be present. In addition, provider characteristics and details about provider-patient relationships, which likely affect guideline-concordant care, were not available. Also, we did not have data on persistently low renal function or albuminuria, which may have led to misclassification of CKD status.
In summary, prevalences of guideline-concordant CKD outcomes are low, though prevalence of BP control has recently increased. In addition, individual awareness of CKD does not appear to be associated with outcomes consistent with guideline concordant care. This suggests that CKD awareness may be important, but insufficient for achievement of optimal CKD care. Greater confidence in our metric to measure CKD awareness is necessary prior to fully embracing these results. Nevertheless, our findings challenge us to better understand the implications of CKD awareness in order to develop educational interventions aimed at improving health outcomes among individuals with CKD.
We thank the participants and staff of NHANES.
None of the authors have any financial disclosures or conflicts of interest to report.
DST was supported by the American Kidney Fund Clinical Scientist in Nephrology Fellowship and currently receives support from Award Number KL2RR024130 from the National Center for Research Resources. C.Y.H. is partially supported by K24DK92291 and N.R.P. is partially supported by grant K24DK02643, both from the National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Md., USA. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.
This work was previously presented in poster format at the American Society of Nephrology Kidney Week, in Denver, Colo., USA in November 2010.
Delphine S. Tuot
1001 Potrero Avenue
Bldg 100, Room 342
Box 1341, San Francisco, CA 94110 (USA)
Tel. +1 415 206 3784, E-Mail Delphine.firstname.lastname@example.org
Received: November 21, 2011
Accepted: December 17, 2011
Published online: January 27, 2012
Number of Print Pages : 7
Number of Figures : 1, Number of Tables : 2, Number of References : 25
American Journal of Nephrology
Vol. 35, No. 2, Year 2012 (Cover Date: February 2012)
Journal Editor: Bakris G. (Chicago, Ill.)
ISSN: 0250-8095 (Print), eISSN: 1421-9670 (Online)
For additional information: http://www.karger.com/AJN