Abstract
Background
Both reduced glomerular filtration rate and increased urine albumin excretion, markers of chronic kidney disease (CKD), are associated with increased risk of atherosclerotic cardiovascular disease (ASCVD). However, CKD is not recognized as an ASCVD risk equivalent by most lipid guidelines. Statin medications, especially when combined with ezetimibe, significantly reduce ASCVD risk in patients with nondialysis-dependent CKD. Unless physicians recognize the heightened ASCVD risk in this population, statins may not be prescribed in the absence of clinical cardiovascular disease or diabetes, a recognized ASCVD risk equivalent. We examined statin use in adults with nondialysis-dependent CKD and examined whether the use differed in the presence of clinical ASCVD and diabetes.
Methods
This study ascertained statin use from pharmacy dispensing records during fiscal years 2012 and 2013 from the US Department of Veterans Affairs Healthcare System. The study included 581 344 veterans aged ≥50 years with nondialysis-dependent CKD Stages 3–5 with no history of kidney transplantation or dialysis. The 10-year predicted ASCVD risk was calculated with the pooled risk equation.
Results
Of veterans with CKD, 62.1% used statins in 2012 and 55.4% used statins continuously over 2 years (2012–13). Statin use in 2012 was 76.2 and 75.5% among veterans with CKD and ASCVD or diabetes, respectively, but in the absence of ASCVD, diabetes or a diagnosis of hyperlipidemia, statin use was 21.8% (P < 0.001). The 10-year predicted ASCVD risk was ≥7.5% in 95.1% of veterans with CKD, regardless of diabetes status.
Conclusions
Statin use is low in veterans with nondialysis-dependent CKD in the absence of ASCVD or diabetes despite high-predicted ASCVD risk. Future studies should examine other populations.
Keywords: cardiovascular disease, chronic kidney disease, epidemiology, prevention, statin medication
INTRODUCTION
Nondialysis-dependent chronic kidney disease (CKD) affects over 500 million people worldwide [1]. The majority of CKD is attributed to high blood pressure and diabetes [1], strong risk factors for atherosclerotic cardiovascular disease (ASCVD), which includes coronary heart disease, stroke, peripheral arterial disease and cardiovascular mortality. For older adults, CKD Stages 3–5 defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 is highly prevalent and may affect one in every three adults aged ≥70 years [2]. However, only a small proportion of individuals with CKD will progress to end-stage renal disease and initiate dialysis or receive a kidney transplant [3–5], while the majority will die from nonkidney-related causes, mainly ASCVD. The risk of ASCVD, including cardiovascular mortality, increases almost exponentially as eGFR declines <60 mL/min/1.73 m2 [4, 6–9]. Thus, ASCVD prevention deserves strong emphasis in the overall clinical management of CKD [10, 11].
Data from randomized trials among individuals with nondialysis-dependent CKD have shown that statins reduce the risk for ASCVD [12–15]. In fact, only blood pressure reduction [16] and use of statin medications [12–15] have been shown to reduce ASCVD risk in adults with CKD, yet few studies have examined factors associated with statin utilization in this population. In this study, we examined statin use during fiscal years (FYs) 2012–13 (e.g. FY 2012 is from 1 October 2011 through 20 September 2012) in a national sample of US veterans aged ≥50 years with nondialysis-dependent CKD Stages 3–5, defined as an eGFR <60 mL/min/1.73 m2, and receiving care at US Veterans Affairs (VA) healthcare facilities. We focused on this age group because current guidelines recommend statin medications for adults with nondialysis-dependent CKD aged ≥50 years [17]. Factors associated with statin use such as presence of ASCVD and diabetes mellitus were also examined, and 10-year predicted ASCVD risk in adults with CKD in the absence of ASCVD was calculated. We hypothesized that statin utilization is low in adults with nondialysis CKD in the absence of ASCVD or diabetes despite high 10-year predicted ASCVD risk.
MATERIALS AND METHODS
Study population
Using data from the national VA Healthcare System databases, we identified a total of 793 274 veterans in FY 2012 with an eGFR <60 mL/min/1.73 m2 calculated from a patient’s last available creatinine measurement from the outpatient laboratory file in FY 2012 using the CKD-EPI equation [18]. Veterans with any dialysis procedure codes (n = 17 458) or transplantation diagnosis codes (n = 730) during FYs 2012–13 were excluded from the analyses. We also excluded veterans <50 years old (n = 9560) and also those with a diagnosis of cancer or death prior to 31 December 2013 (n = 174 111). The analytical files were then linked with the US Renal Data System (USRDS) 2013 database to identify veterans receiving maintenance dialysis or who received kidney transplantation not previously identified from hospital codes. With this linkage, an additional 10 071 veterans who received dialysis were identified and excluded from the analyses. The final study cohort included 581 344 veterans with Stages 3–5 nondialysis-dependent CKD. CKD stages were determined by eGFR values: Stages 3A (eGFR 59–45 mL/min/1.73 m2), 3B (44–30 mL/min/1.73 m2), 4 (29–15 mL/min/1.73 m2) and 5 (<15 mL/min/1.73 m2). The study was approved by the Edward Hines Jr VA Hospital Institutional Review Board.
Statin use
The VA Health System includes VA pharmacies, which dispense medications to the veterans either directly or via mail. After an initial prescription, veterans must then request the VA pharmacy to dispense additional medications to continue medication use. We used pharmacy statin medication dispensing as a surrogate measure of statin use and also ascertained non-VA medications reported by the veterans. Medications dispensed within the VA system were obtained from the Managerial Cost Accounting (MCA) National Data Extracts that contain medications names and medications supply durations. Another source of data, the Corporate Data Warehouse (CDW), contains information about use of medications obtained outside the VA system. Veterans were asked to self-report these medications obtained from non-VA pharmacies and then the VA healthcare providers recorded this information into the electronic medical record (EMR). Supply duration for medications obtained outside the VA is not available in the CDW. Veterans were followed from the beginning of FY 2012 through the end of FY 2013 to ascertain continued statin use. Since medication supply durations were only available for medications acquired from the VA pharmacies, descriptive statistics about supply duration of statins and treatment overlap of statins with other immunosuppressant or cholesterol-lowering agents were limited to drugs dispensed from VA pharmacies. To describe potential statin adherence, the percentage of days covered with statins was calculated using previously described methods [19]. Multiple statin use was defined as switching from one statin to another one during FY 2012.
We also obtained Medicare Part D ‘Slim’ files from the VA Information Resource Center in order to capture medications dispensed from non-VA Medicare-reimbursed pharmacies for veterans aged ≥65 years enrolled in the Medicare Part D program. After the age of 65 years, veterans can opt to obtain their medications through non-VA pharmacies by enrolling in the Medicare Part D Program. We conducted sensitivity analysis for veterans ≥65 years to compare the rate of statins acquired from Medicare-reimbursed pharmacies and non-VA statin medications reported by the veterans in this age group.
Covariates
Covariate selection was based on prior research regarding factors associated with statin use [20]. Demographic factors, comorbidities and geographic residence were ascertained in FY 2011 from the CDW, Medical SAS datasets and the VA vital status file. The demographic characteristics included age, race (black versus nonblack) and ethnicity (Hispanic versus non-Hispanic). Comorbidities were ascertained using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnoses and procedure codes in VA healthcare utilization records, with dates of service during FY 2011 and included ASCVD, hyperlipidemia and diabetes. ASCVD was defined as the presence of acute coronary syndrome, angina, coronary artery disease, heart failure, myocardial infarction, peripheral arterial disease, stroke, transient ischemic attack or coronary revascularization. To account for unobserved differences in treatment approaches and preferences across geographical regions, we identified the US Census Division where each patient lived at the time of diagnosis. We also examined treatment overlap of statins with other immunosuppressant medications or nonstatin cholesterol-lowering agents among patients receiving the statins from VA pharmacies.
The 10-year predicted ASCVD risk was calculated in a subset of 244 651 veterans aged 50–79 years without a history of ASCVD for whom the ASCVD pooled cohort risk equation was applicable [21]. Current tobacco use was obtained from the Health Factors data in the CDW. Cholesterol values, including low-density lipoprotein and high-density lipoprotein (HDL) cholesterol, were obtained from MCA Lab Results data. Systolic blood pressure (SBP) measured during an outpatient clinic visit was obtained from the vital signs file in the CDW. The most recent results for these measurements were identified during FY 2012 and included in the calculation. Complete data on ASCVD risk score variables were available during FY 2012 in 108 961 (44.5%) [21]. To address the missing variables, sensitivity analyses were conducted by replacing all missing ASCVD risk variables with either values reflecting low- to moderate-risk profile (total cholesterol = 200 mg/dL, HDL = 60 mg/dL, SBP = 120 mmHg and no current tobacco use) or a high-risk profile (total cholesterol = 239 mg/dL, HDL = 40 mg/dL, SBP =135 mmHg and current tobacco use).
Statistical analysis
The characteristics of veterans with nondialysis-dependent CKD in FY 2012 were examined by CKD stage. Using descriptive statistics, statin use in FY 2012 and FY 2012–13 was examined by CKD stage, and statin use in FY 2012 was examined in the presence of diabetes, ASCVD and diagnosis of hyperlipidemia. Logistic regression analyses were conducted to examine the odds of statin use in FY 2012 and continued statin use during FY 2012 and FY 2013 by demographic factors, CKD stage, ASCVD comorbidities, diabetes status and statin type with adjustment for covariates. Missing values for demographic characteristics were analyzed as a separate missing category. Analyses of continued statin use were restricted to statin users during FY 2012 (N = 360 908). For the statistical tests used in this study, all reported P-values were two-sided, and statistical significance was defined as α = 0.05. Analyses were conducted using Stata MP Software, version 12.1 (StataCorp LP, College Station, TX, USA).
RESULTS
A total of 581 344 veterans with an eGFR <60 mL/min/1.73 m2 from outpatient laboratory records during FY 2012 with no history of kidney transplantation or dialysis were identified. The majority of veterans with nondialysis-dependent CKD (66.3%) had Stage 3A, with 27.0, 6.0 and 0.7% having CKD Stages 3B, 4 and 5, respectively (Table 1). Overall, 97.0% of the veterans with CKD were male and 57.5% were ≥70 years of age. Race was nonblack in 81.9% and ethnicity was Hispanic in 2.7%. A diagnosis of ASCVD, diabetes and hyperlipidemia was noted in 34.4, 36.9 and 74.6% of the veterans with nondialysis-dependent CKD, respectively; these three conditions were absent in 15.7% of the veterans with nondialysis-dependent CKD.
Table 1.
Demographic and medical characteristics of US veterans with nondialysis-dependent CKD in 2012 by CKD stage (n = 581 344)
| Characteristics |
CKD
|
||||
|---|---|---|---|---|---|
| Stage 3A | Stage 3B | Stage 4 | Stage 5 | Total | |
| n = 385 505 (66.3%) | n = 157 211 (27.0%) | n = 34 788 (6.0%) | n = 3840 (0.7%) | n = 581 344 | |
| Male (%) | 96.9 | 97.3 | 97.1 | 96.5 | 97.0 |
| Age (%), years | |||||
| 50–59 | 5.8 | 3.34 | 5.1 | 11.8 | 5.1 |
| 60–69 | 30.6 | 20.8 | 21.2 | 36.2 | 27.4 |
| 70–79 | 30.0 | 27.4 | 24.5 | 26.2 | 29.0 |
| ≥80 | 33.5 | 48.5 | 49.2 | 25.8 | 28.5 |
| Race (%) | |||||
| Nonblack | 82.0 | 82.2 | 78.7 | 81.0 | 81.9 |
| Black | 10.3 | 10.0 | 13.9 | 15.2 | 10.5 |
| Missing | 7.7 | 7.8 | 7.4 | 3.7 | 7.6 |
| Ethnicity (%) | |||||
| Non-Hispanic | 90.8 | 90.7 | 90.2 | 92.5 | 90.8 |
| Hispanic | 2.7 | 2.6 | 3.3 | 4.3 | 2.7 |
| Missing | 6.5 | 6.7 | 6.5 | 3.2 | 6.5 |
| Residence (%) | |||||
| Urban | 65.2 | 66.1 | 67.0 | 77.8 | 65.6 |
| Rural | 28.1 | 28.1 | 26.8 | 13.6 | 27.9 |
| Missing | 6.7 | 5.8 | 6.1 | 8.6 | 6.4 |
| ASCVD and diabetesa (%) | |||||
| No ASCVD or diabetes | 47.5 | 38.4 | 36.0 | 41.1 | 44.3 |
| Diabetes | 34.2 | 41.4 | 45.1 | 47.4 | 36.9 |
| ASCVD | 32.0 | 39.1 | 40.5 | 28.8 | 34.4 |
| History of hyperlipidemia (%) | 74.2 | 75.6 | 75.2 | 73.6 | 74.6 |
| No history of hyperlipidemia, ASCVD or diabetes (%) | 16.7 | 13.6 | 13.2 | 15.4 | 15.7 |
ASCVDs including acute coronary syndrome, angina, coronary artery disease, myocardial infarction, peripheral arterial disease, stroke, transient ischemic attack and coronary revascularization. ASCVD, diabetes and hyperlipidemia are based on ICD-9 diagnosis codes.
The column percent of ASCVD and diabetes adds up to >100% because of veterans with overlapping ASCVD and diabetes conditions.
The overwhelming majority of statins were obtained from VA pharmacies (97.2% in FY 2012 and 97.4% in FY 2013). Overall adherence appeared high among statin users. The median number of days covered was 98.6% [interquartile range (IQR): 74.0–98.5%]. Among veterans without a diagnosis of diabetes, ASCVD or hyperlipidemia, the median number of days covered with a statin was somewhat lower at 82.2% (IQR: 49.3–98.6%). Figure 1 shows statin use during FY 2012 by CKD stage. Statin use ranged from 64.3% for CKD Stage 3B to 58.4% for Stage 5. Continued statin use during FY 2012 through the end of FY 2013 was similar for CKD Stages 3A, 3B and 4, ranging from 54.6% for Stage 3A to 57.4% for Stage 3B, while 49.7% of veterans with CKD Stage 5 showed continued statin use during FY 2012–13 (Figure 1).
FIGURE 1:
Statin use in veterans with nondialysis-dependent CKD by CKD stage, fiscal years (FYs) 2012–13, n = 581 344.
Statin use by CKD stage and by diagnosis of ASCVD, diabetes and hyperlipidemia is shown in Figure 2. Statin use was the lowest in veterans without a diagnosis of ASCVD, diabetes or hyperlipidemia (21.8%), while the highest use was noted in those with a diagnosis of ASCVD (76.9%) or diabetes (75.5%) (P < 0.001). The adjusted odds of statin use among adults aged ≥50 years with nondialysis-dependent CKD by demographic factors and comorbid conditions are shown in Table 2. Compared with Stage 3A, odds of statin use were slightly higher among veterans with CKD Stage 3B [odds ratio (OR) = 1.08, 95% confidence interval (CI) 1.07–1.10] and Stage 4 (OR = 1.06, 95% CI 1.03–1.08), whereas odds of statin use were lower among veterans with Stage 5 CKD (OR = 0.85, 95% CI 0.79–0.91). Odds of statin use were approximately 5-fold higher among veterans with a diagnosis of hyperlipidemia (95% CI 5.03–5.22), 7-fold higher among veterans with a diagnosis of diabetes (95% CI 7.51–7.83) and 10-fold higher with a diagnosis of ASCVD (95% CI 10.42–10.83) compared with the absence of hyperlipidemia, diabetes or ASCVD diagnoses, respectively. Simvastatin was the most commonly prescribed statin in FY 2012 (58.0%) (Figure 3). A total of 16.6% utilized both statin and nonstatin cholesterol-lowering agents, of whom the majority was taking a statin plus niacin (47.3%). Only 5.5% of patients were taking both statins and immunosuppressants.
FIGURE 2:
Percentage of veterans with statin use in FY 2012 by CKD stage and comorbidities, n = 581 344.
Table 2.
Factors associated with statin use in veterans with nondialysis-dependent CKD (n = 581 344)
| Characteristics | Odds of statin use in 2012 (n = 581 344) | Odds of continued statin use in 2013 among statin users in 2012 (n = 360 908) |
|---|---|---|
| Adjusteda OR (95% CI) | Adjusteda OR (95% CI) | |
| Gender | ||
| Female | 1.0 (Referent) | 1.0 (Referent) |
| Male | 1.10** (1.06–1.14) | 1.36** (1.28–1.44) |
| Age, years | ||
| 50–59 | 1.0 (Referent) | 1.0 (Referent) |
| 60–69 | 1.34** (1.30–1.38) | 1.25** (1.18–1.32) |
| 70–79 | 1.20** (1.16–1.22) | 0.98 (0.93–1.04) |
| ≥80 | 0.96* (0.94–0.99) | 0.82** (0.77–0.86) |
| Race | ||
| Nonblack | 1.0 (Referent) | 1.0 (Referent) |
| Black | 1.06** (1.04–1.08) | 1.07** (1.03–1.11) |
| Ethnicity | ||
| Non-Hispanic | 1.0 (Referent) | 1.0 (Referent) |
| Hispanic | 1.04* (1.00–1.07) | 1.04 (0.97–1.11) |
| Residence | ||
| Urban | 1.0 (Referent) | 1.0 (Referent) |
| Rural | 1.00 (0.99–1.02) | 1.06** (1.04–1.09) |
| Stages | ||
| 3A | 1.0 (Referent) | 1.0 (Referent) |
| 3B | 1.08** (1.07–1.10) | 1.00 (0.97–1.02) |
| 4 | 1.06** (1.03–1.08) | 0.93** (0.89–0.97) |
| 5 | 0.85** (0.79–0.91) | 0.66** (0.58–0.74) |
| Comorbidities | ||
| None | 1.0 (Referent) | 1.0 (Referent) |
| Hyperlipidemia without diabetes or ASCVD | 5.13** (5.03–5.22) | 1.35** (1.30–1.41) |
| Diabetes without ASCVD | 7.67** (7.51–7.83) | 1.59** (1.52–1.67) |
| ASCVD | 10.62** (10.42–10.83) | 1.67** (1.60–1.74) |
| Statin type | ||
| Rosuvastatin | 1.0 (Referent) | |
| Simvastatin | 0.62** (0.60–0.64) | |
| Pravastatin | 0.43** (0.41–0.45) | |
| Atorvastatin | 0.31** (0.29–0.34) | |
| Lovastatin | 0.46** (0.43–0.50) | |
| Fluvastatin | 0.47** (0.34–0.64) | |
| Pitavastatin | 0.01** (0.003–0.04) | |
| Multiple statins | 0.93* (0.88–0.98) | |
| Other cholesterol-lowering agent | 1.0 (Referent) | 1.0 (Referent) |
| No use | ||
| Use | 1.15** (1.13–1.17) | 0.97 (0.94–1.00) |
| Immunosuppressant | ||
| No use | 1.0 (Referent) | 1.0 (Referent) |
| Use | 1.18** (1.15–1.21) | 1.21** (1.15–1.27) |
Analyses accounted for VA and non-VA medications. aAdjusted for variables listed in table. ASCVD including acute coronary syndrome, angina, coronary artery disease, myocardial infarction, peripheral arterial disease, stroke, transient ischemic attack and coronary revascularization.
P < 0.05 and ** P < 0.001.
FIGURE 3:
Type of statin use among veterans with nondialysis-dependent CKD.
Results from the sensitivity analysis incorporating Medicare Part D data for veterans aged ≥65 years enrolled in the Medicare Part D program showed somewhat higher statin use when we accounted for statins acquired from Medicare-reimbursed pharmacies in addition to statins dispensed from VA pharmacies and non-VA statin medications reported by the veterans. Statin use was 72.7% when we accounted for statins acquired from all three sources versus 62.2% when we only accounted for statins dispensed from VA pharmacies and non-VA statin medications reported by the veterans. However, the overall pattern of statin use by CKD stage and comorbidities remained similar to the main results after accounting for statins acquired from Medicare-reimbursed pharmacies. Statin use was again lowest (41.7%) in veterans aged ≥65 years without a diagnosis of ASCVD, diabetes or hyperlipidemia versus 78.3% in veterans aged ≥65 years with those conditions after accounting for statins acquired from Medicare-reimbursed pharmacies, statins dispensed from VA pharmacies and non-VA statin medications reported by the veterans.
The associations between demographic factors and comorbid conditions with continued statin use in FY 2012–13 were similar to those associated with statin use in FY 2012 (Table 2). Statin type was also associated with continued statin use. Compared with rosuvastatin use, other statins such as simvastatin, pravastatin and atorvastatin were all associated with lower odds of continued statin use (Table 2). The 10-year predicted ASCVD risk for veterans aged 50–79 years without history of ASCVD in the presence of diabetes is presented in Table 3. Overall, 95.1 and 89.9% of veterans had a 10-year predicted risk for ASCVD ≥7.5 and ≥10%, respectively, regardless of diabetes status.
Table 3.
ASCVD risk in veterans aged 50–79 years with nondialysis-dependent CKD and without a history of ASCVD by diabetes status in 2012 (n = 244 651)
| Subgroups of veterans | Observations with complete data for ASCVD risk calculations (% nonmissing) | ASCVD risk (%) | Percent of veterans with ASCVD risk (%) |
||
|---|---|---|---|---|---|
| Excluding missing (%) | Replacing missing with low- to moderate-risk profile (%) | Replacing missing with moderate- to high-risk profile (%) | |||
| With diabetes (n = 88 377) | 40 353 (45.7) | ≥7.5 | 96.8 | 97.0 | 98.2 |
| ≥10.0 | 92.9 | 93.1 | 95.8 | ||
| ≥20.0 | 61.1 | 62.0 | 71.9 | ||
| Without diabetes (n = 156 274) | 78 608 (50.3) | ≥7.5 | 94.3 | 93.5 | 96.0 |
| ≥10.0 | 88.3 | 87.1 | 91.8 | ||
| ≥20 | 50.3 | 49.3 | 59.5 | ||
| Total | 118 961 (48.6) | ≥7.5 | 95.1 | 94.8 | 96.8 |
| ≥10.0 | 89.9 | 89.3 | 93.2 | ||
| ≥20.0 | 54.0 | 54.0 | 64.0 | ||
In the low- to moderate-risk profile, the missing variables were populated with the following values: 200 for total cholesterol, 60 for HDL, 120 for SBP and No for current tobacco use.
In the moderate- to high-risk profile, the missing variables were populated with the following values: 239 for total cholesterol, 40 for HDL, 135 for SBP and Yes for current tobacco use.
DISCUSSION
In the population of 581 344 US veterans aged ≥50 years with Stages 3–5 nondialysis-dependent CKD receiving care at US VA healthcare facilities, overall statin use during FY 2012 and 2013 was ∼60%, consistent with previous studies examining statin use in adults with CKD [20, 22]. However, among veterans with nondialysis-dependent CKD in the absence of ASCVD, diabetes or a diagnosis of hyperlipidemia, which comprised of 15.7% of the veteran population with CKD, statin utilization was extremely low. Even after accounting for statins acquired from Medicare-reimbursed pharmacies in veterans aged ≥65 years enrolled in the Medicare Part D program, this statistic remained suboptimal. These findings hold strong clinical and public health implications because decreased eGFR heightens risk for cardiovascular disease, regardless of diabetes status [3, 4], and statin use reduces ASCVD risk in this population [12–15]. Moreover, the predicted ASCVD risk exceeded 10% in almost 90% of these adults with nondialysis-dependent CKD in the absence of diabetes or established ASCVD, findings consistent with a previous analysis of 4726 US adults aged 50–79 years with Stages 3–5 nondialysis-dependent CKD [20]. Odds of statin use were also significantly lower in veterans with nondialysis-dependent CKD Stage 5 as compared with those with CKD Stage 3A. This lower rate of statin use may be due to the perceived risks of statin utilization in advanced CKD or lack of proven benefit of statin medications in adults receiving maintenance dialysis.
The statin type was significantly associated with continued statin use during FY 2012–13 with nonrosuvastatin use associated with lower odds of continued use. During the time period from 2012 to 2013, rosuvastatin was not on the VA formulary and clinicians had to submit nonformulary requests to the hospital pharmacist to prescribe this drug. Thus, patients prescribed this drug may have received additional attention from clinicians, which may explain the higher continued use of this drug. In FY 2012, simvastatin was the most frequently used statin in our study. Simvastatin combined with ezetimibe was used in the largest randomized controlled trial of statins versus placebo for ASCVD outcomes in patients with nondialysis-dependent CKD [13], which may explain the higher use of simvastatin in veterans with nondialysis-dependent CKD. In October of 2012, atorvastatin became available on the VA pharmacy formulary, which may have influenced the conversion of nonformulary statins to atorvastatin. We also found that 16.6% of veterans with nondialysis-dependent CKD Stages 3–5 with statin use were also prescribed other cholesterol-lowering agents including niacin and fenofibrate. We do not have information on the clinical indications for using combined lipid-lowering medication therapy in these patients. Most current guidelines do not recommend combining a statin with fenofibrate in patients with nondialysis CKD due to the high risk of adverse events [23].
The limitations of the study include lack of direct measures of statin use. This study utilized pharmacy dispensing of statin medications as a surrogate measure of statin use. In the VA system, if a provider prescribes or renews a medication, then the prescription will be automatically dispensed to the patient and there will be at least one record of the prescription during the year. The majority of statin users had a 12-month statin supply dispensed during FY 2012 and FY 2013. However, we did not have information as to whether the dispensed statin was actually taken regularly as prescribed. Patients receiving chronic dialysis were identified by procedure codes for dialysis and by linking the data with the USRDS. All veterans receiving any dialysis were excluded, but this may have led to the exclusion of a small number of veterans with CKD who received acute dialysis during the study period. CKD was defined as an eGFR value <60 mL/min/1.73 m2 calculated from a patient’s last available creatinine measurement and this definition could lead to the inclusion of individuals with acute kidney injury. However, the misclassification of CKD would likely be nondifferential with regard to statin use and would not alter the overall findings of low statin use in adults with nondialysis-dependent CKD, especially in the subgroup without diabetes or ASCVD. Approximately 10.5% of veterans aged ≥65 years with nondialysis-dependent CKD acquired statins from non-VA pharmacies, and this medication use was not recorded in their VA EMR. Thus, we may have not captured all statin use. However, even after we accounted for statin use captured by both non-VA and VA pharmacies, statin utilization was lowest in veterans with CKD in the absence of ASCVD, diabetes or a diagnosis of hyperlipidemia. Another limitation is the study time period. Statin utilization may have changed after Kidney Disease: Improving Global Outcomes (KDIGO) published its guideline for lipid management in adults with CKD in 2013 [17]. However, this study provides information on baseline statin use contemporaneous with publication of the KDIGO lipid management guideline. Future studies should examine temporal trends in statin utilization and examine other populations with CKD.
The strengths of this study include large number of adults with nondialysis-dependent CKD and use of data from a health system that provides medications at low cost; thus, statin use will be less influenced by insurance status and other socioeconomic factors. Previous studies have shown generally equal or often better quality of care at VA facilities compared with non-VA facilities in the USA [24, 25]. Thus, the finding of low statin utilization in adults with CKD in the absence of ASCVD or diabetes is probably not unique to veterans receiving care at VA Medical Centers. Because most veterans in this study were male, the findings may not be applicable to women.
In conclusion, statin use appears low in US veterans with nondialysis-dependent CKD in the absence of clinical ASCVD or diabetes, yet 10-year predicted ASCVD risk exceeds 10% in the overwhelming majority of this population. Our findings suggest a need for education efforts to increase statin use in adults with nondialysis-dependent CKD.
FUNDING
This work was funded in part by the Department of Veterans Affairs, Office of Research and Development, Health Services Research and Development HSR&D project IIR 07-165-2. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.
CONFLICT OF INTEREST STATEMENT
The authors declare that they have no financial or nonfinancial competing interests. Data were presented as an oral presentation at the American Society of Nephrology Meeting in San Diego, CA 2016 (D.J.L).
REFERENCES
- 1. Mills KT, Xu Y, Zhang W. et al. A systematic analysis of worldwide population-based data on the global burden of chronic kidney disease in 2010. Kidney Int 2015; 88: 950–957 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Coresh J, Selvin E, Stevens LA. et al. Prevalence of chronic kidney disease in the United States. JAMA 2007; 298: 2038–2047 [DOI] [PubMed] [Google Scholar]
- 3. Keith DS, Nichols GA, Gullion CM. et al. Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 2004; 164: 659–663 [DOI] [PubMed] [Google Scholar]
- 4. Foley RN, Murray AM, Li S. et al. Chronic kidney disease and the risk for cardiovascular disease, renal replacement, and death in the United States Medicare population, 1998 to 1999. J Am Soc Nephrol 2005; 16: 489–495 [DOI] [PubMed] [Google Scholar]
- 5. Newsome BB, McClellan WM, Allison JJ. et al. Racial differences in the competing risks of mortality and ESRD after acute myocardial infarction. Am J Kidney Dis 2008; 52: 251–261 [DOI] [PubMed] [Google Scholar]
- 6. Matsushita K, Coresh J, Sang Y. et al. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diab Endocrinol 2015; 3: 514–525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Matsushita K, van der Velde M, Astor BC. et al. ; Chronic Kidney Disease Prognosis Consortium. 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 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lee M, Saver JL, Chang KH. et al. Level of albuminuria and risk of stroke: systematic review and meta-analysis. Cerebrovasc Dis 2010; 30: 464–469 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Rahman M, Ford CE, Cutler JA. et al. Long-term renal and cardiovascular outcomes in Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) participants by baseline estimated GFR. Clin J Am Soc Nephrol 2012; 7: 989–1002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Sarnak MJ, Bloom R, Muntner P. et al. KDOQI US commentary on the 2013 KDIGO Clinical Practice Guideline for Lipid Management in CKD. Am J Kidney Dis 2015; 65: 354–366 [DOI] [PubMed] [Google Scholar]
- 11. Sarnak MJ, Levey AS, Schoolwerth AC. et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003; 108: 2154–2169 [DOI] [PubMed] [Google Scholar]
- 12. Barylski M, Nikfar S, Mikhailidis DP. et al. Statins decrease all-cause mortality only in CKD patients not requiring dialysis therapy–a meta-analysis of 11 randomized controlled trials involving 21, 295 participants. Pharmacol Res 2013; 72: 35–44 [DOI] [PubMed] [Google Scholar]
- 13. Baigent C, Landray MJ, Reith C. et al. The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial. Lancet 2011; 377: 2181–2192 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Strippoli GFM, Navaneethan SD, Johnson DW. et al. Effects of statins in patients with chronic kidney disease: meta-analysis and meta-regression of randomised controlled trials. BMJ 2008; 336: 645–651 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Upadhyay A, Earley A, Lamont JL. et al. Lipid-lowering therapy in persons with chronic kidney disease: a systematic review and meta-analysis. Ann Intern Med 2012; 157: 251–262 [DOI] [PubMed] [Google Scholar]
- 16. Cheung AK, Rahman M, Reboussin DM. et al. Effects of intensive BP control in CKD. J Am Soc Nephrol 2017; 28: 2812–2823 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kidney Disease: Improving Global Outcomes (KDIGO) Work Group. KDIGO Clinical Practice Guideline for Lipid Management in Chronic Kidney Disease. Kidney Int 2014; 85: 1309–9 [Google Scholar]
- 18. Inker LA, Schmid CH, Tighiouart H. et al. Estimating glomerular filtration rate from serum creatinine and cystatin C. N Engl J Med 2012; 367: 20–29 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Molnar MZ, Gosmanova EO, Sumida K. et al. Predialysis cardiovascular disease medication adherence and mortality after transition to dialysis. Am J Kidney Dis 2016; 68: 609–618 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Colantonio LD, Baber U, Banach M. et al. Contrasting Cholesterol Management Guidelines for Adults with CKD. J Am Soc Nephrol 2015; 26: 1173–1180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Goff DC, Lloyd-Jones DM, Bennett G. et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014; 63 (25 Pt B): 2935–2959 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Schneider MP, Hubner S, Titze SI. et al. Implementation of the KDIGO guideline on lipid management requires a substantial increase in statin prescription rates. Kidney Int 2015; 88: 1411–1418 [DOI] [PubMed] [Google Scholar]
- 23. Markossian T, Burge N, Ling B. et al. Controversies regarding lipid management and statin use for cardiovascular risk reduction in patients with CKD. Am J Kidney Dis 2016; 67: 965–977. [DOI] [PubMed] [Google Scholar]
- 24. Borzecki AM, Christiansen CL, Loveland S. et al. Trends in the inpatient quality indicators: the Veterans Health Administration experience. Med Care 2010; 48: 694–702 [DOI] [PubMed] [Google Scholar]
- 25. Weeks WB, West AN, Rosen AK. et al. Comparing measures of patient safety for inpatient care provided to veterans within and outside the VA system in New York. Qual Saf Health Care 2008; 17: 58–64 [DOI] [PubMed] [Google Scholar]