Advertisement
American Journal of Kidney Diseases

Physician Utilization, Risk-Factor Control, and CKD Progression Among Participants in the Kidney Early Evaluation Program (KEEP)

      Background

      Chronic kidney disease (CKD) is a well-known risk factor for cardiovascular mortality, but little is known about the association between physician utilization and cardiovascular disease risk-factor control in patients with CKD. We used 2005-2010 data from the National Kidney Foundation's Kidney Early Evaluation Program (KEEP) to examine this association at first and subsequent screenings.

      Methods

      Control of risk factors was defined as control of blood pressure, glycemia, and cholesterol levels. We used multinomial logistic regression to examine the association between participant characteristics and seeing a nephrologist after adjusting for kidney function and paired t tests or McNemar tests to compare characteristics at first and second screenings.

      Results

      Of 90,009 participants, 61.3% had a primary care physician only, 2.9% had seen a nephrologist, and 15.3% had seen another specialist. The presence of 3 risk factors (hypertension, diabetes, and hypercholesterolemia) increased from 26.8% in participants with CKD stages 1-2 to 31.9% in those with stages 4-5. Target levels of all risk factors were achieved in 7.2% of participants without a physician, 8.3% of those with a primary care physician only, 9.9% of those with a nephrologist, and 10.3% of those with another specialist. Of up to 7,025 participants who met at least one criterion for nephrology consultation at first screening, only 12.3% reported seeing a nephrologist. Insurance coverage was associated strongly with seeing a nephrologist. Of participants who met criteria for nephrology consultation, 406 (5.8%) returned for a second screening, of whom 19.7% saw a nephrologist. The percentage of participants with all risk factors controlled was higher at the second screening (20.9% vs 13.3%).

      Conclusion

      Control of cardiovascular risk factors is poor in the KEEP population. The percentage of participants seeing a nephrologist is low, although better after the first screening. Identifying communication barriers between nephrologists and primary care physicians may be a new focus for KEEP.

      Index Words

      Chronic kidney disease (CKD) is a well-known risk factor for cardiovascular mortality and morbidity.
      • Go A.S.
      • Chertow G.M.
      • Fan D.
      • et al.
      Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.
      • Matsushita K.
      • van der Velde M.
      • Astor B.C.
      • et al.
      Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis.
      Cardiovascular disease (CVD) risk factors, such as hypertension, diabetes, and dyslipidemia, are highly prevalent and poorly controlled in patients with CKD.
      • Collins A.J.
      • Foley R.N.
      • Herzog C.
      • et al.
      Excerpts from the US Renal Data System 2009 Annual Data Report.
      Recent reports suggest that of patients with an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2, only 37% of those with known hypertension achieved blood pressure control to a level <130/80 mm Hg,
      • Peralta C.A.
      • Hicks L.S.
      • Chertow G.M.
      • et al.
      Control of hypertension in adults with chronic kidney disease in the United States.
      and low-density lipoprotein cholesterol level was within the normal range for 17.9%.
      • Collins A.J.
      • Foley R.N.
      • Herzog C.
      • et al.
      Excerpts from the US Renal Data System 2009 Annual Data Report.
      Most people with early-stage CKD (eGFR >60 mL/min/1.73 m2 with established proteinuria) are managed exclusively by primary care providers, with rates of nephrologist comanagement increasing as CKD progresses.
      • Bayliss E.A.
      • Bhardwaja B.
      • Ross C.
      • et al.
      Multidisciplinary team care may slow the rate of decline in renal function.
      • Hemmelgarn B.R.
      • Zhang J.
      • Manns B.J.
      • et al.
      Nephrology visits and health care resource use before and after reporting estimated glomerular filtration rate.
      • Sprangers B.
      • Evenepoel P.
      • Vanrenterghem Y.
      Late referral of patients with chronic kidney disease: no time to waste.
      The National Kidney Foundation's Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines recommend referral to and/or comanagement by nephrologists for patients with CKD stage 4, macroalbuminuria, hyperkalemia (potassium >5.5 mEq/L), or resistant hypertension or for patients at increased risk of CKD progression.
      • Castro A.F.
      • Coresh J.
      CKD surveillance using laboratory data from the population-based National Health and Nutrition Examination Survey (NHANES).
      National Kidney Foundation
      K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: evaluation, classification and stratification.
      National Kidney Foundation
      K/DOQI Clinical Practice Guidelines on Hypertension and Antihypertensive Agents in Chronic Kidney Disease.
      Timely nephrologist referral has been associated with improved outcomes, including delayed progression to end-stage renal disease, decreased mortality before hemodialysis therapy initiation, and improved first-year survival on hemodialysis therapy.
      • Black C.
      • Sharma P.
      • Scotland G.
      • et al.
      Early referral strategies for management of people with markers of renal disease: a systematic review of the evidence of clinical effectiveness, cost-effectiveness and economic analysis.
      • Chan M.R.
      • Dall A.T.
      • Fletcher K.E.
      • et al.
      Outcomes in patients with chronic kidney disease referred late to nephrologists: a meta-analysis.
      However, little is known about the interplay of physician utilization, CVD risk-factor control, and kidney disease progression in people screened for CKD.
      We used data from the Kidney Early Evaluation Program (KEEP), a community-based health screening program that enrolls participants at high risk of kidney disease, to: (1) assess CVD risk-factor control and physician utilization at baseline, (2) determine predictors of nephrology consultation in participants with identified indications for consultation or referral, and (3) explore CKD progression, CVD risk-factor control, and physician utilization in participants with recurrent KEEP screenings.

      Methods

      KEEP Screening Procedures

      KEEP is a free community-based health screening program that targets populations at high risk of kidney disease. KEEP recruitment methods have been described previously.
      • Brown W.W.
      • Peters R.M.
      • Ohmit S.E.
      • et al.
      Early detection of kidney disease in community settings: the Kidney Early Evaluation Program (KEEP).
      • Jurkovitz C.
      • Qiu Y.
      • Wang C.
      • et al.
      The Kidney Early Evaluation Program (KEEP): program design and demographic characteristics of the population.
      Eligible participants are 18 years or older with self-reported diabetes or hypertension or a first-degree relative with diabetes, hypertension, or kidney disease. People with kidney transplants or receiving regular dialysis therapy are excluded. After providing informed consent, participants complete the screening questionnaire, which consists of sociodemographic information, personal and family health history, smoking status, and information about participant primary care and specialty physicians. Height, weight, blood pressure, plasma glucose, microalbuminuria, and albumin-creatinine ratio (ACR) are measured. Blood samples are drawn from consenting participants and sent to a central laboratory.

      Study Population

      Because lipid measurements at KEEP screenings started in 2005, we limited our study population to participants enrolled in 2005-2010 for whom measurements of eGFR and albuminuria and information about diabetes, hypertension, and cholesterol were available. Because measurement of low-density lipoprotein cholesterol was not available until 2008, we used total cholesterol level to assess hypercholesterolemia.

      Definition of Variables

      Physicians

      Participants who had seen a physician in the past year were considered to have a physician; those not meeting this time criterion were considered not to have a physician. A primary care practitioner was defined as a family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant. Seeing a nephrologist was defined as nephrologist consultation/care with or without a primary care practitioner or another specialist (cardiologist or endocrinologist).

      Comorbid Conditions

      Diabetes was defined as a history of diabetes (self-report or retinopathy), use of diabetes medications, or newly diagnosed diabetes (fasting blood glucose ≥126 mg/dL or nonfasting blood glucose ≥200 mg/dL) in the absence of self-report or medication use. Hypertension was defined as history of hypertension (self-report), use of hypertension medications, or newly diagnosed hypertension
      • Chobanian A.V.
      • Bakris G.L.
      • Black H.R.
      • et al.
      Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.
      defined as systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥80 mm Hg for persons with a history of diabetes or CKD; otherwise, systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg. Hypercholesterolemia was defined as receiving medication for high cholesterol level or total cholesterol level >200 mg/dL.
      CVD was defined as self-reported history of heart angina, heart attack, heart bypass surgery, heart angioplasty, stroke, heart failure, abnormal heart rhythm, or coronary heart disease. Body mass index was calculated as weight (in kilograms) divided by height (in meters) squared.

      Kidney Function

      Serum creatinine was measured and calibrated to the Cleveland Clinic Research Laboratory as previously described.
      • Stevens L.A.
      • Stoycheff N.
      Standardization of serum creatinine and estimated GFR in the Kidney Early Evaluation Program (KEEP).
      GFR was estimated using the CKD Epidemiology Collaboration (CKD-EPI) equation.
      • Levey A.S.
      • Stevens L.A.
      • Schmid C.H.
      • et al.
      A new equation to estimate glomerular filtration rate.
      Microalbuminuria was defined as a spot urine ACR ≥30 mg/g, and macroalbuminuria as ACR >300 mg/g.
      Kidney function stages were defined according to eGFR levels and KDOQI guidelines as follows
      National Kidney Foundation
      K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: evaluation, classification and stratification.
      : normal kidney function, eGFR ≥60 mL/min/1.73 m2 and ACR <30 mg/g; CKD stages 1-2, eGFR ≥60 mL/min/1.73 m2 and ACR ≥30 mg/g; CKD stage 3, eGFR <60 and ≥30 mL/min/1.73 m2; CKD stage 4, eGFR <30 and ≥15 mL/min/1.73 m2; and CKD stage 5, eGFR <15 mL/min/1.73 m2.

      Outcomes

      Control of all risk factors was defined as blood pressure control (systolic blood pressure <130 mm Hg and diastolic blood pressure <80 mm Hg if history of diabetes or CKD; otherwise, systolic blood pressure <140 mm Hg and diastolic blood pressure <90 mm Hg), blood glucose control (fasting blood glucose <126 mg/dL, nonfasting blood glucose <200 mg/dL, and hemoglobin A1c <7%), and cholesterol control (≤200 mg/dL).
      In addition to CKD stage 4 or higher, possible indications for nephrology consultation/referral were macroalbuminuria and risk factors for progression, such as type 2 diabetes with microalbuminuria in patients with eGFR <60 mL/min/1.73 m2.
      • Castro A.F.
      • Coresh J.
      CKD surveillance using laboratory data from the population-based National Health and Nutrition Examination Survey (NHANES).
      Castro et al
      • Castro A.F.
      • Coresh J.
      CKD surveillance using laboratory data from the population-based National Health and Nutrition Examination Survey (NHANES).
      use diabetic retinopathy as a marker of CKD progression in patients with CKD stage 3, but we could not because of inconsistency in its collection in KEEP; we used diabetes with eGFR <60 mL/min/1.73 m2 instead.
      Likewise, we could not use hyperkalemia because it is not assessed in KEEP. Because medication and detailed clinical information are not collected, we could not infer about the presence of resistant hypertension.

      Statistical Analysis

      We used the Cochran-Armitage test of trend to analyze the distribution of participant characteristics according to CKD stages and χ2 tests to evaluate the univariate association between type of physician and risk factors. We used logistic regression to examine the independent association between participant characteristics and all risk-factor control (dependent variable) and multinomial logistic regression for the independent association between participant characteristics and seeing a nephrologist (dependent variable) after adjusting for kidney function. Seeing a nephrologist was compared with seeing another physician or with not seeing a physician. To avoid decreasing the number of records used in the model because of missing data, we created an unknown category for each variable with missing data. Finally, we used paired t tests for continuous variables or McNemar tests for categorical variables to compare participant characteristics at first and second screening.
      Data were analyzed using SAS, version 9.1 (www.sas.com).

      Results

      Participant Population

      A total of 101,439 participants were enrolled in KEEP between 2005 and 2010. Exclusion of participants who had undergone kidney transplant or were receiving hemodialysis (n = 272) and those with missing values for albuminuria, eGFR, hypertension, diabetes, or cholesterolemia (n = 11,158) resulted in a final cohort for analysis of 90,009.
      Of 90,009 participants, 77.2% had no CKD, 8.0% had CKD stages 1-2, 13.9% had stage 3, and 0.9% had stages 4-5 (Table 1). Approximately one-fifth of the study population had not seen a physician in the last year; in the entire cohort, 61.3% had a primary care physician only, 2.9% had seen a nephrologist, and 15.3% had seen another specialist. Of participants with CKD stages 4-5, only 35.3% had seen a nephrologist.
      Table 1Characteristics of KEEP Participants, 2005-2010
      CKD
      AllNoneStages 1-2Stage 3Stages 4-5P
      Test of trend.
      No.90,00969,4927,16612,527824
      Medical care
       No physician20.522.720.09.49.7<0.001
       Primary care
      Family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant.
      only
      61.362.659.256.737.1<0.001
       Nephrologist with or without primary care2.91.63.47.735.3<0.001
       Other specialists
      Cardiologist or endocrinologist.
      with or without primary care
      15.313.017.426.217.8<0.001
      Mean age (y)56.353.755.770.270.7<0.001
      Age ≥65 y31.323.729.571.670.4<0.001
      Men32.031.832.831.937.60.1
      Race/ethnicity
       White51.749.743.167.161.9<0.001
       African American31.031.936.922.825.0<0.001
       Native American2.01.93.31.72.20.5
       Other15.416.516.88.510.9<0.001
       Hispanic12.413.513.75.97.0<0.001
      Any insurance79.077.274.990.786.4<0.001
      Education ≤12 y40.538.645.347.553.1<0.001
      Smoking (prior or current)37.136.141.340.042.9<0.001
      Family history
       Kidney disease16.816.918.415.719.00.2
       Hypertension81.182.179.876.572.8<0.001
       Diabetes54.755.257.550.249.8<0.001
      History of CVD26.923.530.342.149.9<0.001
      Mean BMI (kg/m2)30.330.231.630.029.6<0.001
      BMI ≥30 kg/m244.644.151.543.840.80.1
      Risk factors
       Hypertension81.378.486.893.196.0<0.001
       Diabetes33.529.847.944.653.8<0.001
       Hypercholesterolemia57.055.357.366.260.4<0.001
       3 risk factors
      Hypertension (self-reported history of hypertension, use of antihypertensive medications, or measured systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥80 mm Hg for persons with a history of diabetes or CKD; otherwise, systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg), diabetes (self-reported history of diabetes, retinopathy, or fasting blood glucose ≥126 mg/dL or nonfasting blood glucose ≥200 mg/dL in the absence of self-report of medicine use), and hypercholesterolemia (receiving medication for high cholesterol level or total cholesterol level >200 mg/dL).
      18.215.526.827.631.9<0.001
       2 risk factors43.942.744.150.447.5<0.001
       1 risk factor only29.331.723.320.419.5<0.001
      All risk factors controlled
      In participants with at least one risk factor. Denominator: all participants with hypertension, diabetes, or hypercholesterolemia, as defined.
      8.48.76.08.59.00.02
      Note: Unless otherwise indicated, continuous variables are given as means; categorical variables are shown as percentages. Included KEEP participants with nonmissing values for eGFR and albuminuria and information about diabetes, hypertension, and hypercholesterolemia status.
      Abbreviations: BMI, body mass index; CKD, chronic kidney disease; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; KEEP, Kidney Early Evaluation Program.
      a Test of trend.
      b Family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant.
      c Cardiologist or endocrinologist.
      d Hypertension (self-reported history of hypertension, use of antihypertensive medications, or measured systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥80 mm Hg for persons with a history of diabetes or CKD; otherwise, systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg), diabetes (self-reported history of diabetes, retinopathy, or fasting blood glucose ≥126 mg/dL or nonfasting blood glucose ≥200 mg/dL in the absence of self-report of medicine use), and hypercholesterolemia (receiving medication for high cholesterol level or total cholesterol level >200 mg/dL).
      e In participants with at least one risk factor. Denominator: all participants with hypertension, diabetes, or hypercholesterolemia, as defined.
      Participants with advanced CKD (stages 3-5) were older and more likely to be white, have insurance, and have 12 years or fewer of education.

      CVD Risk-Factor Control and Physician Utilization

      Participants with advanced CKD were more likely to have CVD, hypertension, and hypercholesterolemia (Table 1). The presence of 3 risk factors (hypertension, diabetes, and hypercholesterolemia) was more prevalent with increasing stages of CKD. The rate of control was low; only 8.4% achieved target levels of all risk factors (blood pressure, glycemia, and cholesterolemia). Participants with CKD stages 1-2 were least likely to achieve target levels of all risk factors (6.0%), and those with CKD stages 4-5 were slightly more likely (9.0%). CVD risk-factor control varied little based on physician utilization; 7.2% of participants without a physician, 8.3% of those seeing only a primary care physician, 9.9% of those seeing a nephrologist, and 10.3% of those seeing another specialist achieved target levels of all risk factors. However, nephrologists and specialists were more likely than primary care physicians to see participants with 3 risk factors (28.3% and 30.9%, respectively, vs 17.3%; P < 0.001).
      Results of multivariable analysis confirmed these results (Table 2). After adjusting for demographic and clinical characteristics, participants with CKD stages 1-2 remained 40% less likely to achieve target levels of all risk factors than participants without CKD. CVD risk-factor control was more likely for participants who had seen a physician in the last year than for those who had not, regardless of physician type. Odds ratios were 1.22 (95% confidence interval [CI], 1.14-1.32) for primary care physician, 1.48 (95% CI, 1.35-1.63) for specialist, and 1.52 (95% CI, 1.30-1.63) for nephrologist. Participants with hypertension and hypercholesterolemia were respectively 22% and 70% less likely to achieve target levels, and participants with diabetes were almost 60% more likely.
      Table 2Characteristics Independently Associated With Control of All Risk Factors
      VariableOR (95% CI)P
      No physician1.00 (reference)
      Primary care only1.22 (1.14-1.32)<0.001
      Nephrologist1.52 (1.30-1.77)<0.001
      Specialist
      Cardiologist or endocrinologist.
      1.48 (1.35-1.63)<0.001
      Age0.99 (0.99-1.00)0.004
      Men1.08 (1.03-1.14)0.004
      Race/ethnicity
       White1.00 (reference)
       African American0.93 (0.87-0.98)0.01
       Native American0.71 (0.59-0.87)<0.001
       Other1.19 (1.10-1.29)<0.001
       Hispanic1.12 (1.02-1.22)0.02
      Insurance coverage0.96 (0.89-1.03)0.2
       Unknown (n = 2,765; 3.4%)0.95 (0.82-1.10)0.5
      Education >12 y
      Reference is 12 years or less.
      1.03 (0.97-1.08)0.3
       Unknown (n = 1,093; 1.3%)0.67 (0.52-0.87)0.003
      Family history
       Kidney disease1.03 (0.96-1.10)0.5
        Unknown (n = 5,824; 7.1%)1.04 (0.94-1.16)0.4
       Hypertension1.17 (1.09-1.25)<0.001
        Unknown (n = 6,063; 7.4%)1.72 (1.55-1.92)<0.001
       Diabetes1.01 (0.95-1.07)0.8
        Unknown (n = 5,448; 6.6%)0.99 (0.89-1.12)0.9
      History of CVD1.06 (0.99-1.12)0.07
       Unknown (n = 568; 0.7%)1.39 (1.06-1.84)0.02
      BMI ≥25 kg/m20.97 (0.91-1.03)0.3
       Unknown (n = 886; 1.1%)0.86 (0.67-1.12)0.3
      Hypertension0.88 (0.80-0.95)0.003
      Diabetes1.57 (1.49-1.65)<0.001
      Hypercholesterolemia0.30 (0.29-0.32)<0.001
      CKD
       None1.00 (reference)
       Stages 1-20.60 (0.54-0.67)<0.001
       Stage 30.98 (0.91-1.06)0.6
       Stages 4-50.89 (0.69-1.14)0.3
      Note: OR is for all risk factors controlled. Participants with at least one CVD risk factor (hypertension, diabetes, or hypercholesterolemia), n = 82,313. C index = 0.698.
      Abbreviations: BMI, body mass index; CI, confidence interval; CKD, chronic kidney disease; CVD, cardiovascular disease; OR, odds ratio.
      a Cardiologist or endocrinologist.
      b Reference is 12 years or less.

      Consultation/Referral Indications and Physician Utilization

      A total of 7,025 participants (7.8%) met at least one criterion for nephrology consultation/referral at baseline (Table 3). Of these, 12.3% reported seeing a nephrologist; 50.1%, a primary care physician only; and 29.1%, another specialist. As expected, participants with CKD stages 4-5 (eGFR <30 mL/min/1.73 m2) were most likely to report seeing a nephrologist (35.3%) compared with 11.6% of those with macroalbuminuria and eGFR ≥30 mL/min/1.73 m2 and 12.4% of diabetic participants with microalbuminuria and eGFR of 30-59 mL/min/1.73 m2.
      Table 3Distribution of Medical Care by Referral Criteria
      Medical Care
      AllNo PhysicianPrimary Care Only
      Family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant.
      Other Specialist
      Cardiologist or endocrinologist.
      With or Without Primary Care
      Nephrologist With or Without Other Specialist or Primary CareP
      χ2.
      No.90,00918,46755,18213,7352,625
      Criteria for nephrologist referral
       CKD stages 4-58249.737.117.835.3<0.001
       Macroalbuminuria
      ACR >300 mg/g.
      at CKD stages 1-3
      87915.848.024.611.6<0.001
       Diabetes + microalbuminuria
      ACR of 30-300 mg/g.
      at CKD stage 3
      1,2387.445.634.512.4<0.001
       Diabetes without albuminuria
      ACR <30 mg/g.
      at CKD stage 3
      4,0847.054.630.87.7<0.001
       Any of these criteria7,0258.550.129.112.3<0.001
      Note: Results are row percentages. For example, in participants with CKD stages 4-5, the percentage of participants who have no physician is 9.7. The denominator is number of participants with CKD stages 4-5. Categories are mutually exclusive.
      Abbreviations: ACR, albumin-creatinine ratio; CKD, chronic kidney disease.
      a Family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant.
      b Cardiologist or endocrinologist.
      c χ2.
      d ACR >300 mg/g.
      e ACR of 30-300 mg/g.
      f ACR <30 mg/g.
      Results of the multivariable model assessing the likelihood of seeing a nephrologist versus seeing another physician and versus seeing no physician in participants who met criteria for consultation/referral are listed in Table 4. Because 25.7% of the data were missing, we created an unknown category for each variable with missing data. For both analyses, seeing a nephrologist was associated strongly with decreasing eGFR and increasing albuminuria. After controlling for these factors, several clinical and demographic characteristics also were associated with seeing a nephrologist. Compared with seeing another physician, predictors of seeing a nephrologist were male sex, other race (includes Asians and Pacific Islanders), insurance coverage, more than 12 years of education, family history of kidney disease, and CVD. Participants with diabetes were less likely to see a nephrologist than another physician. Compared with not seeing any physician, the strongest predictor was insurance coverage; this effect was even stronger than effects of eGFR of 30-59 mL/min/1.73 m2 and albuminuria. Other predictors that remained significantly associated with seeing a nephrologist were male sex, more than 12 years of education, family history of kidney disease or hypertension, CVD, and hypertension. Native Americans were more likely to not have a physician.
      Table 4Model Predicting Nephrology Consultation in Participants Who Met Criteria for Referral
      Seeing Nephrologist vs Seeing Another PhysicianSeeing Nephrologist vs Not Seeing a Physician
      VariableOR (95% CI)POR (95% CI)P
      Age0.97 (0.96-0.98)<0.0011.00 (0.99-1.01)0.8
      Men1.45 (1.23-1.70)<0.0011.27 (1.00-1.60)0.05
      Race/ethnicity
       White1.00 (reference)1.00 (reference)
       African American0.94 (0.78-1.14)0.50.76 (0.58-1.00)0.05
       Native American0.73 (0.42-1.27)0.30.48 (0.24-0.97)0.04
       Other1.39 (1.06-1.82)0.020.87 (0.61-1.25)0.5
       Hispanic0.78 (0.55-1.10)0.20.76 (0.49-1.18)0.2
      Insurance coverage1.95 (1.44-2.64)<0.0017.52 (5.33-10.63)<0.001
       Unknown (n = 342; 4.9%)1.97 (1.24-3.14)0.0043.16 (1.82-5.47)<0.001
      Education >12 y
      Reference is 12 years or less.
      1.21 (1.04-1.42)0.011.31 (1.04-1.64)0.02
       Unknown (n = 100; 1.4%)0.68 (0.32-1.47)0.30.47 (0.19-1.19)0.1
      Family history
       Kidney disease1.56 (1.27-1.91)<0.0011.40 (1.03-1.90)0.03
        Unknown (n = 701; 10.0%)1.12 (0.84-1.51)0.40.76 (0.51-1.15)0.2
       Hypertension1.08 (0.88-1.33)0.41.41 (1.05-1.89)0.02
        Unknown (n = 882; 12.6%)1.27 (0.94-1.71)0.11.03 (0.68-1.56)0.9
       Diabetes1.06 (0.89-1.27)0.51.09 (0.84-1.41)0.5
        Unknown (n = 661; 9.4%)1.03 (0.74-1.44)0.90.98 (0.61-1.57)0.9
      History of CVD1.30 (1.11-1.52)0.0011.97 (1.56-2.49)<0.001
       Unknown (n = 42; 0.6%)1.40 (0.51-3.81)0.50.95 (0.27-3.28)0.9
      BMI ≥25 kg/m20.91 (0.74-1.12)0.41.22 (0.91-1.63)0.2
       Unknown (n = 53; 0.8%)0.50 (0.18-1.41)0.20.51 (0.14-1.87)0.3
      Hypertension1.25 (0.85-1.85)0.32.08 (1.28-3.37)0.003
      Diabetes0.71 (0.55-0.92)0.011.17 (0.80-1.69)0.4
      Hypercholesterolemia1.08 (0.92-1.27)0.31.08 (0.86-1.36)0.5
      CKD
       eGFR ≥601.00 (reference)1.00 (reference)
       eGFR of 30-593.35 (2.19-5.15)<0.0013.45 (2.01-5.91)<0.001
       eGFR <3014.61 (9.47-22.51)<0.00110.24 (5.88-17.82)<0.001
       No albuminuria1.00 (reference)1.00 (reference)
       ACR of 30-3001.63 (1.36-1.95)<0.0011.53 (1.16-2.03)0.003
       ACR >3002.19 (1.71-2.81)<0.0011.95 (1.33-2.86)<0.001
      Note: Results from multinomial logistic regression (n = 7,025).
      Abbreviations and definitions: ACR, albumin-creatinine ratio (in mg/g); BMI, body mass index; CI, confidence interval; CKD, chronic kidney disease; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate (in mL/min/1.73 m2); OR, odds ratio.
      a Reference is 12 years or less.

      Physician Utilization and CVD and Kidney Disease Progression Risk-Factor Control at Subsequent Screening

      Of participants with at least one indication for consultation/referral, 406 (5.8%) returned for a second KEEP screening (Table 5). The average interval between screenings was 1.55 years (median, 1.02 years). Compared with participants who met criteria for consultation/referral but did not return (n = 6,619), those who returned were more likely to have a physician and to see a specialist (P = 0.03). They were older (72.3 vs 69.3 years; P < 0.001), more likely to be white (72.7% vs 63.5%; P = 0.001) and to have insurance (92.1% vs 88.1%; P = 0.02), and less likely to smoke (35.2% vs 41.2%; P = 0.02). They were more likely to have CVD risk factors (hypertension, diabetes, and hypercholesterolemia; 64.5% vs 55.6%; P < 0.001) and CKD stage 3 (91.6% vs 80.8%; P < 0.001) and less likely to have CKD stages 4-5 (6.9% vs 11.9%; P = 0.002) and macroalbuminuria (5.4% vs 12.6%; P < 0.001). The proportion of participants who saw a nephrologist increased from 11.6% to 19.7% (P < 0.001) between screenings (Table 5). Participants were more likely to have all 3 CVD risk factors at the return visit (72.9% vs 64.5% at baseline; P < 0.001), largely due to more diagnoses of hypercholesterolemia; however, the percentage of participants with all risk factors controlled was higher at the second than at the first screening (20.9% vs 13.3%; P = 0.002).
      Table 5Risk-Factor Control and CKD Progression in Participants Who Met Criteria for Nephrologist Referral and Returned for a Second KEEP Screening
      KEEP Screening
      Met Criteria for Nephrologist ReferralFirstSecondP
      Paired t test or McNemar test.
      No.7,025406406
      Physician care
       No physician8.55.23.50.2
       Primary care only
      Family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant.
      50.149.042.60.01
       Nephrologist12.311.619.7<0.001
       Other specialist
      Cardiologist or endocrinologist.
      and primary care
      29.134.234.21.0
      Mean age (y)69.472.373.9<0.001
      Age ≥65 y69.980.383.3<0.001
      Men34.232.832.8
      Values were the same for both screenings.
      Race/ethnicity
       White62.872.772.7
      Values were the same for both screenings.
       African American24.517.517.5
      Values were the same for both screenings.
       Native American2.61.01.0
      Values were the same for both screenings.
       Other10.18.98.9
      Values were the same for both screenings.
       Hispanic7.36.26.2
      Values were the same for both screenings.
      Any insurance88.191.792.30.5
      Education ≤12 y50.645.744.70.4
      Smoking (former or current)40.933.633.30.8
      Family history of kidney disease15.017.117.1
      Values were the same for both screenings.
      History of CVD46.650.450.60.9
      Mean BMI (kg/m2)31.330.930.70.1
      BMI ≥30 kg/m251.550.149.40.6
      Hypertension94.695.396.10.5
      Diabetes90.396.897.00.7
      Hypercholesterolemia64.168.778.1<0.001
      Presence of risk factors
      Hypertension (self-reported history of hypertension, use of antihypertensive medications, or measured systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥80 mm Hg for persons with a history of diabetes or CKD; otherwise, systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg), diabetes (self-reported history of diabetes, retinopathy, or fasting blood glucose ≥126 mg/dL or nonfasting blood glucose ≥200 mg/dL in the absence of self-report of medicine use), and hypercholesterolemia (receiving medication for high cholesterol level or total cholesterol level >200 mg/dL).
       355.664.572.9<0.001
       238.231.725.40.01
       15.83.71.70.02
      All risk factors controlled
      In participants with at least one risk factor. Denominator: all participants with hypertension, diabetes, or hyperlipidemia, as defined.
      10.213.320.90.002
      CKD stages 1-27.21.51.51.0
      CKD stage 381.191.689.70.1
      Criteria for nephrologist referral
       CKD stages 4-511.76.98.90.1
       Macroalbuminuria
      ACR >300 mg/g.
      12.55.46.20.5
       Diabetes + microalbuminuria
      ACR of 30-300 mg/g.
      at CKD stage 3
      17.618.519.00.8
       Diabetes without albuminuria
      ACR <30 mg/g.
      at CKD stage 3
      58.169.266.00.1
      Note: Unless otherwise indicated, values are percentages.
      Abbreviations: ACR, albumin-creatinine ratio; BMI, body mass index; CKD, chronic kidney disease; CVD, cardiovascular disease; KEEP, Kidney Early Evaluation Program.
      a Paired t test or McNemar test.
      b Family practice physician, internist, obstetrician/gynecologist, gerontologist, nurse practitioner, or physician assistant.
      c Cardiologist or endocrinologist.
      d Values were the same for both screenings.
      e Hypertension (self-reported history of hypertension, use of antihypertensive medications, or measured systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥80 mm Hg for persons with a history of diabetes or CKD; otherwise, systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg), diabetes (self-reported history of diabetes, retinopathy, or fasting blood glucose ≥126 mg/dL or nonfasting blood glucose ≥200 mg/dL in the absence of self-report of medicine use), and hypercholesterolemia (receiving medication for high cholesterol level or total cholesterol level >200 mg/dL).
      f In participants with at least one risk factor. Denominator: all participants with hypertension, diabetes, or hyperlipidemia, as defined.
      g ACR >300 mg/g.
      h ACR of 30-300 mg/g.
      i ACR <30 mg/g.

      Discussion

      We investigated CVD risk-factor control and physician utilization in KEEP participants and in the subset who returned for a subsequent screening. The major findings are: (1) generally poor risk-factor control and only modest improvement with advancing CKD, (2) low likelihood of nephrologist encounter despite clinical indications for consultation/referral at earlier CKD stages, (3) higher likelihood of a nephrologist visit after the first screening, and (4) improved CVD risk-factor control in returning participants.
      Hypertension, diabetes, and hyperlipidemia are highly prevalent in patients with end-stage renal disease or CKD.
      • Go A.S.
      • Chertow G.M.
      • Fan D.
      • et al.
      Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.
      • Collins A.J.
      • Foley R.N.
      • Herzog C.
      • et al.
      Excerpts from the US Renal Data System 2009 Annual Data Report.
      Of National Health and Nutrition Examination Survey (NHANES) participants with eGFR <60 mL/min/1.73 m2, only 37% of those with known hypertension had normal blood pressure.
      • Peralta C.A.
      • Hicks L.S.
      • Chertow G.M.
      • et al.
      Control of hypertension in adults with chronic kidney disease in the United States.
      Likewise, both diabetes and hyperlipidemia control are poor in patients with CKD.
      • Collins A.J.
      • Foley R.N.
      • Herzog C.
      • et al.
      Excerpts from the US Renal Data System 2009 Annual Data Report.
      Secondary analyses of large clinical trials of statins for primary prevention of cardiovascular events show a beneficial effect in patients with CKD
      • Kendrick J.
      • Shlipak M.G.
      • Targher G.
      • et al.
      Effect of lovastatin on primary prevention of cardiovascular events in mild CKD and kidney function loss: a post hoc analysis of the Air Force/Texas Coronary Atherosclerosis Prevention Study.
      • Ridker P.M.
      • MacFadyen J.
      • Cressman M.
      • et al.
      Efficacy of rosuvastatin among men and women with moderate chronic kidney disease and elevated high-sensitivity C-reactive protein: a secondary analysis from the JUPITER (Justification for the Use of Statins in Prevention—an Intervention Trial Evaluating Rosuvastatin) trial.
      ; however, physicians have been reluctant to prescribe statins for fear of secondary effects
      • Thompson P.D.
      • Clarkson P.
      • Karas R.H.
      Statin-associated myopathy.
      and due to lack of efficacy in randomized controlled trials of hemodialysis patients.
      • Wanner C.
      • Krane V.
      • Marz W.
      • et al.
      Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis.
      As expected, we found that the prevalence of CVD risk factors increased with kidney disease severity. Risk-factor control is low (8.4%) in the KEEP population, possibly explaining the high rates of cardiovascular events and death reported previously.
      • McCullough P.
      • Li S.
      • Jurkovitz C.
      • et al.
      CKD and cardiovascular disease in screened high-risk volunteer and general populations: the Kidney Early Evaluation Program (KEEP) and National Health and Nutrition Examination Survey (NHANES) 1999-2004.
      • McCullough P.A.
      • Jurkovitz C.T.
      • Pergola P.E.
      • et al.
      Independent components of chronic kidney disease as a cardiovascular risk state Results from the Kidney Early Evaluation Program (KEEP).
      Interestingly, participants with CKD stages 4-5 seem to have slightly better control of risk factors than those with less advanced CKD, possibly due to a larger proportion reporting nephrologist care. In the overall KEEP population, risk-factor control does not seem to depend on type of physician seen. However, nephrologists and other specialists are more likely to see patients with high levels of comorbidity, and controlling risk factors in such practice settings might be more difficult.
      Almost 8% of KEEP participants met criteria for nephrologist consultation/referral. This probably is an underestimate because we could not include participants with resistant hypertension or hyperkalemia. In NHANES, Castro and Coresh
      • Castro A.F.
      • Coresh J.
      CKD surveillance using laboratory data from the population-based National Health and Nutrition Examination Survey (NHANES).
      found in patients with CKD stage 3 that 18.6% met one of these referral criteria. Another possible reason for our lower prevalence is that we did not limit our analysis to participants with CKD stage 3.
      Only 12.3% of participants who met any referral criterion reported seeing a nephrologist. This low referral rate may be related to the low CKD awareness (10.0%) consistently reported in KEEP.
      • Tamura M.K.
      • Anand S.
      • Li S.
      • et al.
      Comparison of CKD awareness in a screening population using the Modification of Diet in Renal Disease (MDRD) Study and CKD Epidemiology Collaboration (CKD-EPI) equations.
      The referral rate increases to 19.7% at the second screening, which does not strongly support the notion that awareness increases nephrologist utilization. The decision to refer to a nephrologist depends on physician and participant factors, and one of the major goals of KEEP is to improve awareness of CKD in both these groups.
      Primary care practitioner awareness of the KDOQI guidelines is a critical factor in nephrology referral decisions. Although distinguishing awareness from motivation is challenging, several investigators have attempted to assess knowledge of these guidelines among physicians. Navaneethan et al
      • Navaneethan S.D.
      • Kandula P.
      • Jeevanantham V.
      • et al.
      Referral patterns of primary care physicians for chronic kidney disease in general population and geriatric patients.
      recently found that only 36.5% of primary care practitioners were aware of CKD guidelines and only 31.8% used CKD stages for referral. In a cross-sectional survey of internists, geriatricians, and nephrologists, regarding referral of older patients, investigators reported that 100% of surveyed nephrologists, 31.3% of internists, and 57.1% of geriatricians were aware of the KDOQI guidelines related to referral.
      • Campbell K.H.
      • Sachs G.A.
      • Hemmerich J.A.
      • et al.
      Physician referral decisions for older chronic kidney disease patients: a pilot study of geriatricians, internists, and nephrologists.
      A subsequent study showed that primary care physicians with more than 10 years in practice were least likely to recommend referral of patients with CKD but more likely to express a desire for collaborative care, yet the differences were small (89% vs 82%).
      • Boulware L.E.
      • Troll M.U.
      • Jaar B.G.
      • et al.
      Identification and referral of patients with progressive CKD: a national study.
      • Diamantidis C.J.
      • Powe N.R.
      • Jaar B.G.
      • et al.
      Primary care-specialist collaboration in the care of patients with chronic kidney disease.
      General internists who were aware of existing guidelines were 14 times more likely to recommend referral.
      • Boulware L.E.
      • Troll M.U.
      • Jaar B.G.
      • et al.
      Identification and referral of patients with progressive CKD: a national study.
      In our analysis, after adjusting for kidney disease progression, participant factors associated with seeing a nephrologist included male sex, insurance coverage, more than 12 years of education, family history of kidney disease and CVD. Notably, participants with insurance coverage were nearly twice as likely to be referred to a nephrologist as those without insurance, compared with seeing another physician. These results are similar to results reported by other investigators, who found that patient characteristics such as age older than 65 years, female sex, and nonwhite race were significantly associated with nonreferral.
      • Navaneethan S.D.
      • Kandula P.
      • Jeevanantham V.
      • et al.
      Referral patterns of primary care physicians for chronic kidney disease in general population and geriatric patients.
      Although the small group of participants who returned for a second screening seems to be a highly selected population of older participants with better socioeconomic status, only 19.7% reported having seen a nephrologist. Nevertheless, KEEP seems to have been successful in encouraging a nephrology visit because this is a 70% increase from the first screening. KEEP is actively engaged in a longitudinal program, inviting previous participants to return for a repeated examination. These results suggest that rescreening, in addition to focusing on participants with criteria for CKD progression, should focus on the most vulnerable participants (no health insurance, minority race/ethnicity, and low level of education). Finally, a large percentage of KEEP participants who meet criteria for referral have seen a physician in the year preceding the first screening. Although KEEP provides the screening results to consenting participants' physicians, lack of improvement or deterioration remains prevalent at the second screening. Communication barriers between primary care physicians and specialists should be assessed, as should barriers to guideline implementation.
      The definition of CKD based on a single eGFR and ACR measurement, not on measurements over 3 months, is a limitation inherent in the cross-sectional design of KEEP, as is ascertainment of ACR as the only marker of kidney damage. This definition may lead to overestimating CKD prevalence in our study population because some individuals with acute changes in kidney function may have been misclassified. The small number of participants who met criteria for kidney disease progression and returned for a second screening is another serious limitation. Because this is a self-selected group likely highly motivated for care, selection bias may have been introduced, and the improvement in percentage of nephrologist visits and risk-factor control may be overestimated. In addition, because of the small numbers of participants, we could not assess the impact of physician visits on clinical outcomes. However, these results provide insight into the effectiveness of screening regarding participant referral. Finally, we could assess parameters at only the screening and return screening; an analysis including interim data between these visits would likely further elucidate the nature of improvements (or lack thereof) in risk factors.
      In conclusion, we found that a large number of participants met criteria for referral to a nephrologist and that control of cardiovascular risk factors was poor in the KEEP population, but seemed to improve after screening. Socioeconomic status, including insurance coverage, is a major patient-related determinant of nephrology consultation. Although KEEP was effective in increasing the percentage of participants seeing a nephrologist, the rate was low and probably overestimated in our sample. These results also highlight that a large percentage of the population who returned had seen a physician in the year before the second screening. Identifying the communication barriers between nephrologists and primary care physicians may be a new focus for KEEP, particularly with the current emphasis on accountable care organizations and medical home designations.

      Acknowledgements

      The KEEP Investigators are Peter A. McCullough, Adam T. Whaley-Connell, Andrew Bomback, Kerri Cavanaugh, Linda Fried, Claudine Jurkovitz, Mikhail Kosiborod, Samy McFarlane, Rajnish Mehrotra, Keith Norris, Rulan Savita Parekh, Carmen A. Peralta, Georges Saab, Stephen Seliger, Michael Shlipak, Lesley Inker, Manjula Kurella Tamura, John Wang; ex-officio, Bryan Becker, Allan Collins, Nilka Ríos Burrows, Lynda A. Szczech, Joseph Vassalotti; advisory group, George Bakris, Wendy Brown; data coordinating center, Shu-Cheng Chen.
      We thank the participants and staff who volunteered their time to make the KEEP screening a successful event and Chronic Disease Research Group colleagues Shane Nygaard, BA, for manuscript preparation and Nan Booth, MSW, MPH, ELS, for manuscript editing.
      Support: The KEEP is a program of the National Kidney Foundation Inc and is supported by Amgen, Abbott, Siemens, Astellas, Fresenius Medical Care, Genzyme, LifeScan, Nephroceuticals, and Pfizer. Dr Norris receives support from National Institutes of Health grants RR026138 and MD000182 . Dr Whaley-Connell receives support from the Veteran's Affairs Career Development Award-2, National Institutes of Health grant R03AG040638-01 , and American Society of Nephrology-Association of Specialty Professors-National Institute on Aging Development Grant in Geriatric Nephrology.
      Financial Disclosure: The authors declare that they have no other relevant financial interests.

      References

        • Go A.S.
        • Chertow G.M.
        • Fan D.
        • et al.
        Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.
        N Engl J Med. 2004; 351: 1296-1305
        • Matsushita K.
        • van der Velde M.
        • Astor B.C.
        • et al.
        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
        • Collins A.J.
        • Foley R.N.
        • Herzog C.
        • et al.
        Excerpts from the US Renal Data System 2009 Annual Data Report.
        Am J Kidney Dis. 2010; 55 (A426-A427): S1-S420
        • Peralta C.A.
        • Hicks L.S.
        • Chertow G.M.
        • et al.
        Control of hypertension in adults with chronic kidney disease in the United States.
        Hypertension. 2005; 45: 1119-1124
        • Bayliss E.A.
        • Bhardwaja B.
        • Ross C.
        • et al.
        Multidisciplinary team care may slow the rate of decline in renal function.
        Clin J Am Soc Nephrol. 2011; 6: 704-710
        • Hemmelgarn B.R.
        • Zhang J.
        • Manns B.J.
        • et al.
        Nephrology visits and health care resource use before and after reporting estimated glomerular filtration rate.
        JAMA. 2010; 303: 1151-1158
        • Sprangers B.
        • Evenepoel P.
        • Vanrenterghem Y.
        Late referral of patients with chronic kidney disease: no time to waste.
        Mayo Clin Proc. 2006; 81: 1487-1494
        • Castro A.F.
        • Coresh J.
        CKD surveillance using laboratory data from the population-based National Health and Nutrition Examination Survey (NHANES).
        Am J Kidney Dis. 2009; 53: S46-S55
        • National Kidney Foundation
        K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: evaluation, classification and stratification.
        Am J Kidney Dis. 2002; 39: S17-S31
        • National Kidney Foundation
        K/DOQI Clinical Practice Guidelines on Hypertension and Antihypertensive Agents in Chronic Kidney Disease.
        Am J Kidney Dis. 2004; 43: S1-S290
        • Black C.
        • Sharma P.
        • Scotland G.
        • et al.
        Early referral strategies for management of people with markers of renal disease: a systematic review of the evidence of clinical effectiveness, cost-effectiveness and economic analysis.
        Health Technol Assess. 2010; 14: 1-184
        • Chan M.R.
        • Dall A.T.
        • Fletcher K.E.
        • et al.
        Outcomes in patients with chronic kidney disease referred late to nephrologists: a meta-analysis.
        Am J Med. 2007; 120: 1063-1070
        • Brown W.W.
        • Peters R.M.
        • Ohmit S.E.
        • et al.
        Early detection of kidney disease in community settings: the Kidney Early Evaluation Program (KEEP).
        Am J Kidney Dis. 2003; 42: 22-35
        • Jurkovitz C.
        • Qiu Y.
        • Wang C.
        • et al.
        The Kidney Early Evaluation Program (KEEP): program design and demographic characteristics of the population.
        Am J Kidney Dis. 2008; 51: S3-S12
        • Chobanian A.V.
        • Bakris G.L.
        • Black H.R.
        • et al.
        Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.
        Hypertension. 2003; 42: 1206-1252
        • Stevens L.A.
        • Stoycheff N.
        Standardization of serum creatinine and estimated GFR in the Kidney Early Evaluation Program (KEEP).
        Am J Kidney Dis. 2008; 51: S77-S82
        • Levey A.S.
        • Stevens L.A.
        • Schmid C.H.
        • et al.
        A new equation to estimate glomerular filtration rate.
        Ann Intern Med. 2009; 150: 604-612
        • Kendrick J.
        • Shlipak M.G.
        • Targher G.
        • et al.
        Effect of lovastatin on primary prevention of cardiovascular events in mild CKD and kidney function loss: a post hoc analysis of the Air Force/Texas Coronary Atherosclerosis Prevention Study.
        Am J Kidney Dis. 2010; 55: 42-49
        • Ridker P.M.
        • MacFadyen J.
        • Cressman M.
        • et al.
        Efficacy of rosuvastatin among men and women with moderate chronic kidney disease and elevated high-sensitivity C-reactive protein: a secondary analysis from the JUPITER (Justification for the Use of Statins in Prevention—an Intervention Trial Evaluating Rosuvastatin) trial.
        J Am Coll Cardiol. 2010; 55: 1266-1273
        • Thompson P.D.
        • Clarkson P.
        • Karas R.H.
        Statin-associated myopathy.
        JAMA. 2003; 289: 1681-1690
        • Wanner C.
        • Krane V.
        • Marz W.
        • et al.
        Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis.
        N Engl J Med. 2005; 353: 238-248
        • McCullough P.
        • Li S.
        • Jurkovitz C.
        • et al.
        CKD and cardiovascular disease in screened high-risk volunteer and general populations: the Kidney Early Evaluation Program (KEEP) and National Health and Nutrition Examination Survey (NHANES) 1999-2004.
        Am J Kidney Dis. 2008; 51: S38-S45
        • McCullough P.A.
        • Jurkovitz C.T.
        • Pergola P.E.
        • et al.
        Independent components of chronic kidney disease as a cardiovascular risk state.
        Arch Intern Med. 2007; 167: 1122-1129
        • Tamura M.K.
        • Anand S.
        • Li S.
        • et al.
        Comparison of CKD awareness in a screening population using the Modification of Diet in Renal Disease (MDRD) Study and CKD Epidemiology Collaboration (CKD-EPI) equations.
        Am J Kidney Dis. 2011; 57: S17-S23
        • Navaneethan S.D.
        • Kandula P.
        • Jeevanantham V.
        • et al.
        Referral patterns of primary care physicians for chronic kidney disease in general population and geriatric patients.
        Clin Nephrol. 2010; 73: 260-267
        • Campbell K.H.
        • Sachs G.A.
        • Hemmerich J.A.
        • et al.
        Physician referral decisions for older chronic kidney disease patients: a pilot study of geriatricians, internists, and nephrologists.
        J Am Geriatr Soc. 2010; 58: 392-395
        • Boulware L.E.
        • Troll M.U.
        • Jaar B.G.
        • et al.
        Identification and referral of patients with progressive CKD: a national study.
        Am J Kidney Dis. 2006; 48: 192-204
        • Diamantidis C.J.
        • Powe N.R.
        • Jaar B.G.
        • et al.
        Primary care-specialist collaboration in the care of patients with chronic kidney disease.
        Clin J Am Soc Nephrol. 2011; 6: 334-343