Adherence to Statins and LDL-Cholesterol Goal Attainment
Published Online: April 17, 2014
Margaret D. Chi, MPH; Southida S. Vansomphone, PharmD; In-Lu Amy Liu, MS; T. Craig Cheetham, PharmD, MS; Kelley R. Green, PhD; Ronald D. Scott, MD; and Kristi Reynolds, PhD, MPH
Cardiovascular disease is the leading cause of death in the United States with 1 out of 5 deaths attributable specifically to coronary artery disease (CAD). Each year, approximately 635,000 people in the United States will have a new coronary attack and 280,000 will have a recurrence.1 Patients living with CAD have an increased risk of a recurrent coronary event,2-4 which can lead to disability or death. In addition, in 2008, CAD had an estimated total national economic burden of $190.3 billion.1
To reduce the risk of recurrent events, the National Cholesterol Education Program (NCEP) recommends a low-density lipoprotein cholesterol (LDL-C) goal level of less than 100 mg/dL for patients with CAD through the use of lipid-lowering drugs such as hydroxymethylglutaryl-CoA reductase inhibitors (statins),5 and studies have shown that those who have reached this goal have lower rates of recurrence and mortality.6-9 Despite their efficacy, the use of these drugs in CAD patients has been shown to be suboptimal.10-12
Although studies have focused separately on LDL-C levels as well as adherence to statins among CAD patients, little is known about how adherence directly affects lipid levels in this population. We examined the association between LDL-C goal attainment and adherence to statins among CAD patients. By understanding the association of adherence and LDL-C goal attainment status, it will be possible to target the appropriate populations to ultimately improve patient treatment and outcomes.
Study Setting and Population
This study was conducted at Kaiser Permanente Southern California (KPSC), an integrated healthcare system that provides care to approximately 3.5 million members across 10 counties in Southern California. The Institutional Review Board at KPSC approved this study.
The source population was identified from the KPSC CAD Registry, which identifies patients with CADrelated International Statistical Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes from hospitalization and outpatient and inpatient CAD-related ICD-9-CM procedure codes (available upon request). The registry includes “risk equivalent” atherosclerotic cardiovascular diseases including ischemic stroke, peripheral arterial disease, and abdominal aortic aneurysm. Deceased members and those living in skilled nursing facilities are excluded from the registry.
Patients eligible for this study included all adult members 18 years or older in the registry as of May 27, 2010, with 12 months of continuous prior enrollment. Patients were included if they had filled at least 2 statin prescriptions (simvastatin or a combination of simvastatin and ezetimibe) and had at least 1 LDL-C result in this time period. Patients with rhabdomyolysis, creatinine kinase greater than 10,000 IU/L, allergy or intolerance to statins, and those without continuous membership or drug benefits between May 2009 and May 2010, were excluded. A total of 67,100 members were eligible and included in this analysis (Figure 1).
Data Collection and Measures
Information on patient demographics including age, gender, race/ethnicity, and clinical information including laboratory tests, prescription medications, healthcare utilization and comorbidities were extracted from health plan electronic databases. When multiple LDL-C measurements were present, the most recent measurement in the study period was extracted. LDL-C goal attainment was defined according to guidelines by the NCEP5 as less than 100 mg/dL and the optional goal of less than 70 mg/dL for these high-risk patients.
Information on healthcare utilization during the study period included the number of unique prescription medications (based on the generic product identified subclass), the numbers of clinic and emergency department (ED) visits and number of hospital admissions. The Deyo adaptation of the Charlson Comorbidity Index (CCI) was determined using ICD-9-CM diagnosis codes from inpatient and outpatient encounters as an overall measure of disease burden.13
Pharmacy fill data were extracted for each eligible individual and a medication possession ratio (MPR) was calculated looking back 400 days from May 27, 2010, as the sum of the days of supply divided by the total number of days between the first dispense date and May 27, 2010, when the most recent prescription was overdue. If the most recent prescription was not overdue, then MPR was calculated as the sum of the days of supply, excluding the last refill, divided by the total number of days between the first dispense date and last dispense date. The most recent refill was disregarded when the prescription was not overdue to avoid inflating the MPR. Adherence was defined as MPR 80% or greater, as suggested by previous studies.4,14
Descriptive statistics (means and proportions) were used to present patient characteristics categorized by LDLC goal attainment (<100 mg/dL; optional <70 mg/dL). Mean LDL-C levels were calculated according to various levels of adherence. Logistic regression was used to estimate the effect of adherence on LDL-C goal attainment controlling for multiple patient characteristics including adherence to statin medication, age, gender, race/ethnicity, concomitant medication use, utilization of healthcare services, CCI, and hypertension. All statistical analysis was performed using SAS version 9.2 (SAS Institute, Cary, North Carolina).
Characteristics of the study population stratified by LDL-C goal attainment are shown in Table 1. Nearly 86% of the population reached the LDL-C goal of less than 100 mg/dL; 32.4% reached the optional goal of LDL-C less than 70 mg/dL; and 79.8% of the population was adherent to their statin medication (MPR ≥80%). Patients at goal had higher adherence rates (LDL-C <100 mg/dL: 82.1% vs 66.0%; LDL-C <70 mg/dL: 84.3% vs 77.7%) and lower mean LDL-C values (LDL-C <100 mg/dL: 73.4 mg/ dL vs 121.0 mg/dL; LDL-C <70 mg/dL: 57.3 mg/dL vs 91.1 mg/dL) compared with those not at goal. Patients achieving LDL-C of less than 100 mg/dL and LDL-C less than 70 mg/dL were more likely to be taking lower-dose statins and more medications compared with those not at medicagoal. In general, patients at goal were also sicker, with a higher CCI score and higher percentages of diabetes, hypertension, and prior myocardial infarction (MI). Patients achieving LDL-C less than 100 mg/dL had slightly lower utilization of hospitalizations and ED visits in the past year while patients achieving LDL-C less than 70 mg/dL had higher utilization of hospitalizations and ED visits compared with those not at goal.
In this population of patients chronically taking statin medications, increasing adherence as determined by MPR was associated with improved LDL-C levels. LDL-C goal attainment had a positive association with increasing adherence, but goal attainment leveled off once adherence was over 86% (Figure 2). Men were consistently more adherent than women, and adherence increased with age (Figure 3).
Results of the multivariable logistic analysis are presented in Table 2. Patients who were adherent to their statin medication were much more likely to be at LDL-C goal than patients who were not adherent (LDL-C <100 mg/dL: odds ratio [OR], 2.24; 95% confidence interval [CI], 2.13-2.35; LDL-C <70 mg/dL: OR, 1.55; 95% CI, 1.48-1.62). Other patient characteristics associated with LDL-C goal attainment of less than 100 mg/dL included increased age, male sex, and Hispanic and Asian race/ethnicity (compared with whites). Patients prescribed more medications, and who had higher CCI scores, and who had hypertension, were more likely to have LDL-C less than 100 mg/dL. In contrast, black patients with CAD had lower odds of LDL-C less than 100 mg/dL compared with whites. Patients with more hospitalizations and ED visits in the prior year were also less likely to have LDLC of less 100 mg/dL. Results of the multivariable logistic regression analysis using LDL-C goal attainment of less than 70 mg/dL were similar although age and ED visits were no longer significantly associated with LDL-C goal attainment and patients who were hospitalized were more likely to be at LDL-C goal of less than 70 mg/dL.
In this study of patients with CAD in a large integrated healthcare system, 85.9% of patients were at the recommended LDL-C goal of less than 100 mg/dL and 79.8% were adherent to their statin medication. Among patients who were not at LDL-C goal, 66.0% were adherent while 17.9% of patients who were at goal were not adherent (defined as MPR <80%) to their statin medication. Adherence to statin medication, older age, male sex, Hispanic and Asian race/ethnicity, ambulatory utilization, higher CCI scores, and hypertension were associated with a higher likelihood of LDL-C goal attainment.
PDF is available on the last page.