Patient perceptions and use of non‐statin lipid lowering therapy among patients with or at risk for atherosclerotic cardiovascular disease: Insights from the PALM registry

Abstract Background Non‐statin lipid lowering therapies (LLTs) provide additional treatment options for patients. Use patterns and patient perceptions of non‐statin LLT remain incompletely described. Hypothesis The guideline‐recommended statin intensity remains underutilized in patients treated with and without non‐statin LLT. Methods The PALM Registry collected LLT information on patients with or at risk of ASCVD treated at 125 US clinics in 2015. We compared patient perceptions, lipid levels and statin use among patients treated with and without non‐statin LLT. Results Among 7720 patients, 1930 (25.0%) were treated with a non‐statin LLT (1249 fish oil, 417 fibrates, 329 ezetimibe, 196 niacin). Concurrent statin treatment occurred in 73.7%, of which 45.4% were dosed under the guideline‐recommended intensity. Compared with patients on statin alone, patients receiving both a statin and non‐statin LLT (n = 1423) were more likely to be male, white race and to perceive themselves as higher risk of ASCVD compared with their peers (38.5% vs. 34.9%, p = .047). Only 27.4% of patients treated with non‐statin LLT alone perceived themselves at higher risk. Most (75.7%) patients treated with a non‐statin LLT alone reported never being treated with a statin, despite ASCVD in 30.8% of these patients. Among those previously treated with a statin, 59.3% reported being willing to try a statin again. Conclusions Non‐statin LLT is used in one in four patients with or at risk for ASCVD; its use is frequently in place of statin therapy or in the absence of guideline‐recommended statin intensity. More work is needed to establish statins as first line therapy.


| BACKGROUND
The benefits of treatment with statins are well recognized among patients with or at risk for clinical atherosclerotic cardiovascular disease (ASCVD). While non-statin lipid lowering therapies (LLTs), including fish oil, fibrates, ezetimibe and niacin, are also used in clinical practice, randomized trial data have shown mixed results in the ability of these medications to affect clinical outcomes. In the wake of several negative clinical trials, the prescription of non-statin LLT among Medicare beneficiaries decreased between 2007 and 2011, from 20.5% to 18.9% of patients, driven by a marked reduction in the number of patients prescribed ezetimibe. 1

| Study population
The PALM registry is a cross-sectional registry conducted following the release of the 2013 ACC/AHA lipid guidelines, and which uniquely collected patient and clinician perceptions of ASCVD risk and treatment preferences, as well as detailed patient clinical characteristics and core lab lipid measurements. Between May 27, 2015 and November 12, 2015, the PALM registry prospectively enrolled 7938 patients from 140 outpatient cardiovascular, endocrinology and primary care practices across the United States. Full details regarding the design, rationale, inclusion and exclusion criteria for the PALM registry have been previously published. 5 In brief, patients were eligible for enrollment if they were treated with a statin or had an indication for statin therapy under the 2013 ACC/AHA guideline. Prior to enrollment, all patients provided signed informed consent. Each site obtained institutional review board approval for participation.
Starting with the overall population of patients, we sequentially excluded patients who did not have completed chart abstraction (N = 34), those who did not have core lipid sample results available (N = 182) and those missing data on statin use (N = 2). This left a final study population of 7720 patients. Among these patients, 7676 (99.4%) had an identified clinician (N = 453) who completed a provider survey from 125 clinic sites, allowing for the linkage of patient data with clinician characteristics and survey responses.

| Data collection and definitions
At the time of enrollment, medical history, focused on cardiovascular disease history, interventions or risk factors, and sociodemographic information, including race, zip code and insurance payor, were abstracted from each patient's medical record. Current and prior treatment with statin and non-statin LLT were also recorded. Treatment with PCSK9 inhibitors were not collected as this study was conducted prior to widespread availability. Each patient underwent phlebotomy with total cholesterol, direct low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and triglyceride levels measured by a core laboratory at LabCorp (Burlington, NC).
Each patient enrolled in the PALM registry was asked to complete a survey (Table S1) which examined the perception of their own cardiovascular risk relative to their peers, as well as their beliefs about the benefit and risks associated with statin therapies. The survey asked patients to indicate the level to which they agreed or disagreed (strongly agree, agree, neither agree nor disagree, disagree, strongly disagree, do not know/not sure) with statements such as "Statins are safe medications", "I think statins can cause diabetes" and "Statin medications are effective in reducing the risk of heart disease and stroke". For patients with current or prior statin or non-statin LLT use, the survey asked patients to select the reason(s) for taking each medication from a list which included "my 'bad' cholesterol was too high", "to prevent stroke", "to prevent heart attack", among others. The survey also queried prior adverse symptoms related to prior or current statin therapy and interventions attempted to reduce symptoms. Table S1.

Questions and answer choices are shown in
Clinicians treating the enrolled patients were asked to complete a survey querying their specialty, number of years in practice, and whether their clinical practice was associated with a teaching institution. Clinicians were also asked to indicate how they would respond to a hypothetical patient already on atorvastatin 80 mg daily for secondary prevention who still had suboptimal lipid levels (total cholesterol 220 mg/dl, LDL cholesterol 130 mg/dl, and HDL cholesterol 30 mg/dl); potential answer choices included add ezetimibe, add fibrate, add fish oil, add bile acid sequestrant, change to a different statin, refer to lipid specialist, no change, and other.

| Statistical analysis
The proportion of patients treated with each non-statin LLT (fish oil, fibrate, ezetimibe, niacin) was described. Concomitant statin use, as well as the frequency of guideline recommended statin intensity was summarized based on non-statin LLT treatment.
We linked each surveyed clinician to the patients they treated in PALM. The clinician characteristics were compared for patients receiving non-statin LLT treatment versus those who did not. We   When patients treated with a non-statin LLT were asked the reason for treatment, 61.0% reported that a non-statin LLT was prescribed because "bad cholesterol level was too high", 27.3% believed that the non-statin LLT medication was prescribed to "prevent stroke" and 38.9% believed that the non-statin LLT was prescribed to "prevent heart attack". Table 2 shows the patient responses for treatment indication based on non-statin LLT prescribed.

| Patient treated with LLT alone versus no treatment
Among patients not on statin therapy, those treated with a non-statin LLT were more likely to be male, of white race, and less likely to be  Prior literature supports that treatment with non-statin LLTs can modify lipids. 8,9 Given its demonstrated LDL-C lowering and clinical efficacy in secondary prevention patients (IMPROVE-IT trial), guidelines and expert consensus recommend the addition of ezetimibe as the first non-statin LLT. 2,4,10 However, we observed that the majority (64.7%) of non-statin LLT users were treated with fish oil, a medication that, until the recent REDUCE-IT trial with optimized doses of icosapent ethyl, did not have strong evidence toward clinical benefit. 11,12 To date, studies have revealed no clear benefit in the reduction of cardiovascular events with the treatment of fibrates or niacin. [13][14][15] When patients specified the indication for treatment with a non-statin LLT, nearly half (45.5%) of patients treated with niacin and 28.6% of those treated with fibrates reported these were to "reduce the risk of heart attack".
While the rationale for the observed approach cannot be elucidated from the data available in this study, it may be that the majority of patients treated with a non-statin LLT received fish oil related to its availability as an over-the-counter supplement and perceived safety, whereas other non-statin LLTs require a prescription. Patients treated with fish oil or niacin were more likely to be cared for by a cardiologist; cardiologists were also more likely to add a non-statin LLT to the patient with elevated blood lipid levels despite the use of highintensity statin therapy. This may be related to a greater awareness of the ACC/AHA guideline recommendations for use of non-statin LLT among clinicians specializing in this field.
Over half of patients receiving both a statin and non-statin LLT were treated for secondary prevention of ASCVD. This group of patients was more likely to have other high-risk features, such as diabetes and family history of premature ASCVD. They were also more likely to self-report higher risk for a cardiovascular event than their age-and gender-matched peers, compared with patients treated with statins alone. Nearly half of the patients reported that the non-statin LLT was prescribed specifically to prevent cardiovascular events.
While there is now randomized clinical trial data to support the benefit of ezetimibe in reducing cardiovascular events, 3  The guideline recommendations support statins as the cornerstone of LLT but with the use of ezetimibe as an important adjunct to statin therapy for a patient with known or at high risk for ASCVD who had a response to statin therapy that was less than anticipated, or who

| Limitations
We acknowledge several limitations associated with our study. First, core laboratory lipid levels were only obtained at the time of study enrollment reflecting lipid treatment received at that time, thus changes in lipids with addition of LLT cannot be described. Second, we were unable to capture motivations behind the lipid-lowering treatment decisions; some of the patients not treated or undertreated with a statin based on guideline recommendations may have been decisions made after risk discussions between the clinician and patient. Data for additional non-statin lipid lower therapies, including bile acids and red yeast rice were collected but there were an insufficient number of patients on these medications to further evaluate in this analysis. For patients using fish oil as a lipid lowering therapy, additional details surrounding this medicationincluding over the counter versus prescription therapywere not available. Finally, PCSK-9 inhibitors were not yet clinically available for use during the enrollment period for the PALM Registry, therefore our description of non-statin LLTs did not include this class of therapy.

| CONCLUSIONS
One in four patients with or at risk of ASCVD in the PALM Registry was treated with a non-statin lipid lowering therapy. While ezetimibe is an important adjunct to statin therapy, particularly for those patients without the expected response to a statin or who are intolerant to statins, we found that non-statin LLT was frequently used in place of statin therapy. Patients treated with a non-statin LLT alone had lower self-perceived ASCVD risk, and the majority had never received statin therapy. Further work is needed to address patient knowledge gaps in their own ASCVD risk, as well as to establish statins as first line therapy before considering non-statin LLT additions or alternatives.