In patients with cardiovascular (CV) disease (CVD), the initiation of and adherence to evidence-based pharmacotherapy are key in preventing additional CV events. However, in complex healthcare systems, a proportion of these patients may not be receiving optimal CV risk–lowering medications.1 Furthermore, it may be challenging to achieve optimal statin use in underserved populations, such as those at Federally Qualified Health Centers. A recent evaluation of 68 patients with atherosclerotic CVD (ASCVD) in 2 Federally Qualified Health Centers showed that only 67% of men and 63% of women were prescribed a statin.2
A new Medicare Star measure aimed at correcting statin underutilization in the secondary CV prevention population was added to Medicare Part C plans in 2019.3 The Statin Therapy for Patients with Cardiovascular Disease (SPC) measure assesses the percentage of men aged 21 to 75 years and women aged 40 to 75 years with ASCVD who receive at least 1 prescription fill for a moderate- to high-intensity statin therapy during the measurement year. This measure is based on Healthcare Effectiveness Data and Information Set (HEDIS) criteria and is endorsed by the National Committee for Quality Assurance.3
Clinical pharmacists are uniquely positioned to affect positively the achieved health system performance on the SPC measure when directly outreaching to patients with ASCVD about the benefits of statin therapy. A 2018 review article discusses several pharmacist-led interventions to improve a similar Medicare Star measure, the Statin Use in Persons with Diabetes measure.4 Interventions that targeted providers via pharmacotherapy recommendations were inefficient, with only 16.6% to 32.4% of statin recommendations being accepted by providers.4 Many of the interventions included in those studies relied on faxes sent or phone calls to make these recommendations, and the lack of a pharmacist–prescriber relationship might have led to the low response and low acceptance rates seen in these studies.4
Conversely, the use of independent pharmacist prescribing authority in Canada demonstrated a 25% absolute increase in the achievement of a low-density lipoprotein cholesterol (LDL-C) target compared with usual care from their physician and pharmacist (18% vs 43%) in one study, and a 21% relative reduction (an approximate 5.4% absolute reduction) in CV events in another study, by allowing pharmacists to prescribe statins to patients directly.5,6
Because many clinical pharmacists in the United States can prescribe or order medications for patients under a collaborative pharmacy practice agreement or similar protocols, a pharmacist intervention that directly intervenes in the treatment of patients with ASCVD who are not receiving statin therapy may be the most effective method for increasing statin prescribing in a high-risk population and improving the performance related to the SPC measure.
A 2013 systematic review of 5 randomized controlled trials (N = 2568 patients) of interventions by clinical pharmacists to optimize the use of appropriate CV risk–lowering medications (including statins) in patients with CVD showed conflicting conclusions.7 All 5 studies reported on adherence to lipid-lowering therapy, and their results showed a significant difference in adherence in the intervention group in 2 of the 5 studies. The same 2 studies demonstrated a statistically and clinically significant increase in LDL-C goal attainment rate.7
One large study of 30 primary care practices (N = 7586 patients) in the United Kingdom assessed an intensive 12-month intervention by pharmacists who worked 1 day per week in their assigned clinics to deliver the intervention.8 The intervention activities included provider education, support in identifying patients with CVD who were not receiving optimal statin therapy, and pharmacotherapy recommendations.8
The percentage of patients achieving target lipid levels was significantly greater in the intervention group compared with the usual care group (69.5% vs 63.5%, respectively; odds ratio [OR], 1.11; 95% confidence interval [CI], 1-1.23; P = .043).8 The intervention in this study was aimed at making recommendations to primary care providers (PCPs) rather than having clinical pharmacists directly intervene to optimize therapy.
In a study of 253 patients at 4 Veterans Affairs (VA) medical centers, a multidisciplinary, year-long intervention was compared with usual care in patients with acute coronary syndrome after discharge.9 The pharmacist portion of the intervention included medication reconciliation, face-to-face education, voice messages for refills and further education, visits 7 to 10 days after discharge, the provision of pillboxes, and a 1-month phone call to follow-up on new medications and synchronize refills.9
The mean additional time spent over 12 months of intervention totaled 3 hours 51 minutes per patient. At study completion, the intervention group had significantly higher proportions of days covered for statins than the usual care group (93.2% vs 71.3%, respectively; P <.001). No difference, however, was found in the secondary outcome of LDL-C goal attainment over the relatively short study period. This study achieved its results with a time- and labor-intensive intervention.9
It is unknown whether successful clinical pharmacist interventions that increased statin use in other (eg, patients with diabetes, patients without ASCVD) populations are transferrable to patients who meet the SPC measure classification within a safety-net health system. The purpose of our study was to assess the effectiveness of a clinical pharmacist intervention on the percentage of patients with ASCVD within a safety-net health system who are prescribed a moderate- or high-intensity statin therapy. Unlike previous studies, our intervention was designed for real-world application, which allowed for direct patient intervention by a clinical pharmacist and targeted a safety-net population.
This single-center, pre- and postintervention study included patients with ASCVD in managed care plans who were not currently receiving a statin. The study was created to test the feasibility and effectiveness of expanding it to a systemwide initiative. This study was conducted at Denver Health Medical Center, a large, urban, safety-net healthcare system with 10 Federally Qualified Health Center clinics that are also patient-centered medical homes. Clinical pharmacists are located at 7 of these 10 primary care clinics.
Denver Health Medical Plan insures more than 100,000 people, approximately 4600 of whom were Medicare-eligible at the study’s initiation. This study was approved by the Colorado Multiple Institutional Review Board and by Denver Health’s Sponsored Programs and Research Office.
At study initiation in January 2019, Denver Health Medical Plan had not yet received any patient-level SPC measure data from the HEDIS vendor. Therefore, data analysts for the health plan created a report that modeled the HEDIS specifications.
Patients were included if they were men aged 21 to 75 years or women aged 40 to 75 years, received primary care at Denver Health Medical Center, had clinical ASCVD (defined by the SPC measure), and had no current or recently filled prescription for a statin.
Clinical ASCVD is defined by the SPC measure as having a history of at least 1 of the following conditions: inpatient discharge with myocardial infarction in the year before the measurement year, coronary artery bypass graft in any setting in the year before the measurement year, percutaneous coronary intervention or other revascularization in any setting in the year before the measurement year, or at least 1 inpatient or outpatient visit for ischemic vascular disease in the measurement year and the year before the measurement year.10
The study’s exclusion criteria were based on the SPC measure, including hospice patients or patients with any of the following conditions in the past year, including pregnancy, in vitro fertilization, at least 1 dispensing event for clomiphene, end-stage renal disease, cirrhosis, or diagnosis of muscle symptoms (ie, myalgia, myositis, or rhabdomyolysis).10 Additional exclusion criteria included liver function tests ≥3 times the upper limit of normal, active liver disease (not cirrhosis), previous allergy or contraindication to statin therapy (because the SPC measure only assesses statin use in the past year), significant drug interactions, or currently breastfeeding. Vulnerable populations, including prisoners and patients with severe mental illness or mental disabilities, were excluded from the study based on Colorado Multiple Institutional Review Board’s requirements.
The primary intervention for eligible patients was telephone outreach by a clinical pharmacist and the initiation of treatment with a moderate- or high-intensity statin, as appropriate (Table 1). This included dose escalation for patients receiving low-intensity statins (if clinically appropriate) and prescription renewal or refills for patients who were nonadherent to moderate- or high-intensity statins. If the indication for statin therapy was unclear, or if there was an unresolved clinical question, the clinical pharmacist contacted the PCP before contacting the patient.
Telephone interventions were individualized, depending on the patient’s unique clinical situation, but generally included a discussion of ASCVD, the risks for secondary CV events, the role of statins in preventing these events, the importance of statin adherence, medication cost, and a discussion of the potential risks associated with statin use. If a patient was not reached by the second call, documentation of the 2 attempted calls and rationale for the need for a statin were recorded in the electronic health record (EHR) and an EHR message was sent to the patient’s PCP.
If a patient accepted a statin, the drug was prescribed by the clinical pharmacist via a collaborative pharmacy practice agreement, if such an agreement was already in place for the patient, or with a PCP’s cosignature and approval. The PCP was alerted to the statin’s initiation through a message in the EHR and/or was prompted by a cosignature order request. If the patient or the PCP declined the initiation of the statin, the reason for declination was documented.
Charts for patients who received a statin prescription were reviewed postintervention to document if and when the patient picked up the statin (ie, a dispensing event occurred). The proportion of patients meeting the SPC measure classification of receiving statin therapy was compared after the intervention with the proportion that was calculated before the intervention.
All the patients who were contacted for the study were offered the opportunity to consent to participation in the study via a telephonic consent process. Patients were allowed to receive the statin prescription even when opting out of having their data recorded for the study.
The primary outcome was the proportion of Denver Health Medical Center primary care patients who met the SPC measure criteria and who were eligible for a clinical pharmacist phone intervention and had 1 dispensing event for a moderate- or high-intensity statin by the end of the study period.
A dispensing event was determined by review of the EHR, which indicated whether the medication was picked up at a Denver Health Medical Center pharmacy, or by calling the community pharmacy to which the prescription was sent. This outcome was designed to assess the potential impact of the clinical pharmacist’s phone intervention on the SPC Star measure parameter of at least 1 dispensing event for a moderate- to high-intensity statin in the past year.
The secondary outcomes included the proportion of the SPC measure–eligible patients who were reached via telephone within 2 attempts, the total time spent per patient, the proportion of patients reached via telephone for whom a statin was prescribed, and the proportion of patients who picked up their statin prescription within 10 days.
Three preplanned subgroup analyses included female patients, patients with chronic kidney disease (estimated glomerular filtration rate, 15-59 mL/min/1.73 m2 and not on hemodialysis), and patients with heart failure and reduced ejection fraction, as identified on the patient’s problem list and/or on an echocardiogram with a left-ventricular ejection fraction of ≤40%.
Descriptive statistics were used for the primary outcome, as well as to describe the patients’ baseline characteristics, the total time spent per patient by clinical pharmacist, the proportion of eligible patients able to be reached within 2 telephone calls, the proportion of patients reached via telephone for whom a statin was prescribed, and the proportion of patients who picked up a prescribed statin within 10 days. The pre- and postintervention proportions for the subgroup analyses were compared using the chi-square test. If the value for any expected frequency was 5 or less, a 2-tailed Fisher’s exact test was used instead.
Internal data from the Denver Health Medical Plan yielded a total of 84 of approximately 4600 Medicare patients who were potentially eligible for the intervention (Figure). Chart review revealed that 26 patients were identified inappropriately, because they were already prescribed a statin therapy or they did not have ASCVD. An additional 6 patients met 1 or more of the exclusion criteria, and 7 patients were deemed ineligible for outreach (eg, having previously refused statin therapy on multiple occasions). In all, 45 patients were eligible for telephone outreach.
A total of 11 patients required discussion with the PCP before outreach; of these patients, 2 had PCPs who denied the initiation of a statin. One of these 2 PCPs opted to discuss statin therapy with the patient at the next visit, and the other PCP opted for no statin treatment because of the patient’s history of cirrhosis and unclear diagnosis of ASCVD. This left 33 patients to be contacted (Figure).
The mean age of the intervention population was 65.8 years (standard deviation [SD], 6.3 years), and 45.7% of the population was female (Table 2). Only 1 (2.9%) patient was receiving a nonstatin antihyperlipidemic medication (gemfibrozil).
On average, the preintervention LDL-C level was above the 70-mg/dL threshold for secondary prevention (mean, 80.9 mg/dL; SD, 38.1); however, there was wide variation in how recently the laboratory test was drawn. Approximately 51% of the patients needed a statin initiated, and approximately 49% needed assistance with adherence to statin therapy or an increase in statin intensity (Table 2).
After 2 outreach attempts to each patient, 22 (66.7%) patients were reached (Table 3). Of the 35 eligible patients meeting the SPC measure classification who were identified from the Denver Health Medical Plan data set, 20 (57.1%) received a moderate- to high-intensity statin (Table 3).
The mean total time spent per patient was 27.7 (SD, 9) minutes. Of the patients eligible for the telephone intervention, approximately 67% were able to be reached. Of the 22 patients reached, 18 (81.8%) accepted a statin prescription. An additional 4 patients were prescribed a statin by a provider other than the clinical pharmacist. These prescriptions came shortly after the clinical pharmacist was unable to reach a patient and an EHR note and an EHR message to the PCP were created; they were likely an indirect result of the intervention. Of the 22 patients who were prescribed a statin, 16 (72.7%) obtained the statin within 10 days (Table 3).
A subgroup analysis of female patients showed no sex-based difference in terms of patients picking up a statin within 10 days after it was prescribed or the total eligible patients who received a moderate- to high-intensity statin by the end of the study period. There was a nonsignificant difference in the proportion of patients who accepted a statin as a result of the intervention (100% of women vs 63.6% of men; P = .09; Table 4).
The 2 preplanned subgroup analyses of patients meeting the SPC measure classification who had heart failure, reduced ejection fraction, and chronic kidney disease were unable to be conducted because of the low numbers of patients (Table 2).
We evaluated whether telephone outreach by a clinical pharmacist to prescribe statin therapy for patients with ASCVD could impact the performance of the SPC Medicare Star measure in a safety-net population. The higher-than-80% acceptance rate for the intervention in the contacted patients, as well as the more than 50% of patients eligible for intervention who ended up receiving statin therapy, indicate that clinical pharmacists are well-positioned to affect statin use in the secondary CV prevention population positively.
Of note, 4 of the 22 (18.2%) patients who received a statin therapy did not have it prescribed by the clinical pharmacist during the intervention phone call; rather, a provider initiated the prescription shortly after an unsuccessful telephone intervention. This suggests that documentation of a patient’s indication for statin therapy in the medical record may positively influence statin prescribing patterns.
Because the number of interventions for statin adherence are nearly as many as those for the initiation of a statin, this may indicate that the intervention was focused more on restarting statin therapy than on newly starting statin therapy.
A recent large retrospective cohort study of VA patients demonstrated the importance of statin adherence.11 The study assessed all-cause death relative to deciles of statin medication possession ratio (MPR) in 347,104 patients with ASCVD. The cohort with the highest MPR (≥90%) had a significantly lower hazard ratio than even the next most adherent group (70%-89% MPR).11
Persistence with statin therapy, especially in high-risk patients, is key to the prevention of CV events. Although the length and design of our study were not intended to assess statin persistence or adherence, which are not part of the SPC Star measure, the results did show that clinical pharmacists doing such work can identify and start to intervene in the treatment of patients with poor statin adherence.
In early 2019, HEDIS vendor data became available that could be compared with the internal SPC measure report used for this study. The data revealed that the number of patients meeting the SPC measure classification who had 1 dispensing event for a moderate- or high-intensity statin in the past year increased from 58 of 235 (24.7%) patients in January 2019 to 159 of 281 (56.6%) patients in March 2019 (P <.001).
In addition, data from a Centers for Medicare & Medicaid Services (CMS) report showed an improvement in Denver Health Medical Plan’s SPC measure adherence from 79% (3 Star rating) to 88% (5 Star rating) between January 2019 and August 2019. A 5 Star rating is the highest achievable Star rating for a CMS Star measure.
These data, along with the HEDIS vendor report, showed improvement in SPC measure performance during the study period. Although the 20 additional patients who were receiving a moderate- or high-intensity statin as a result of this study did not affect the health plan’s improved performance by themselves, they did demonstrate that telephonic intervention by a clinical pharmacist can be part of a suite of interventions by a health plan to move the needle on the SPC measure. This impact could be more significant going forward if the intervention is continued using the larger population identified in the HEDIS vendor reports.
Not all ambulatory care clinical pharmacists participate in population health work on a day-to-day basis. This may be based on a variety of factors, including time. Therefore, it was important to assess whether the intervention in this study was effective, as well as feasible to accomplish within real-world time constraints.
The average time spent of approximately 28 minutes per patient is lengthy. However, the time spent in our study included the time to obtain consent from patients and perform additional documentation for the study. In a nonresearch setting, these steps would be unnecessary. Although the consent process was not timed separately from the rest of the intervention, it is our estimate that consent took between 5 and 15 minutes. Based on this estimate, a clinical pharmacist doing this type of outreach could expect to spend approximately 20 minutes per patient.
The impetus for the preplanned subgroup analysis of female patients was based on a recent large retrospective cohort study of 88,256 patients with Medicare or with private insurance who had a hospital stay for myocardial infarction.12 The risk ratio (RR) for filling a high-intensity statin prescription was significantly lower in the overall female population versus men (RR, 0.91; 95% CI, 0.90-0.92).12 In addition, female sex was a predictor of lower adherence in the previously described VA study of 347,104 patients (OR, 0.89; 95% CI, 0.84-0.94; P <.001).11
The subgroup analysis in our study ran contrary to these results. We found no significant differences between the sexes in the percentages of patients who eventually received a statin or between the patients who picked up their statin. These differences may reflect the unique nature of a safety-net population compared with the VA population or may be a result of our study’s small sample size.
One limitation of this study was its small sample size, which was the result of the health plan’s internal report inaccurately reflecting the total population of patients who met the SPC measure classification and were eligible for the intervention based on HEDIS criteria. Once the HEDIS data were received, it became apparent that there were substantially more health plan patients in the SPC measure population than initially thought.
The discrepancy in the population size (84 patients from the internal report vs 235 patients from the HEDIS vendor report) is likely a result of multiple factors. It is possible that the lag time from the internal report to the receipt of the HEDIS vendor report allowed for additional members to newly qualify for the SPC measure population (either by plan enrollment or by having a new diagnosis or event).
In addition, ASCVD is a complex diagnosis to define accurately and extract from a database, which may be why some of the 84 patients from the internal SPC measure report turned out to not have ASCVD based on the chart review, whereas other patients who were identified with ASCVD on the HEDIS vendor report did not show up on the internal report.
A natural next step is to confirm the success shown in the smaller internal SPC measure report cohort using the larger population of the HEDIS vendor reports that now are regularly received by the health plan.
The results in this study indicate that the use of clinical pharmacists to initiate, reinitiate, or increase the intensity of statins via telephone is an efficient and effective method for ensuring that patients in a safety-net population who have ASCVD are receiving guideline-recommended statin therapy. Clinical pharmacists should be used in this capacity to improve SPC Medicare Star ratings, as well as to improve the health of this high-risk population.
Author Disclosure Statement
Dr Cornelison, Dr Marrs, and Dr Anderson have no conflicts of interest to report.
- Harrison TN, Scott RD, Cheetham TC, et al. Trends in statin use 2009–2015 in a large integrated health system: pre- and post-2013 ACC/AHA Guideline on Treatment of Blood Cholesterol. Cardiovasc Drugs Ther. 2018;32:397-404.
- Khan NNS, Kelly-Blake K, Luo Z, Olomu A. Sex differences in statin prescribing in diabetic and heart disease patients in FQHCs: a comparison of the ATPIII and 2013 ACC/AHA cholesterol guidelines. Health Serv Res Manag Epidemiol. 2019;6:1-8.
- Centers for Medicare & Medicaid Services. Medicare 2019 Part C & D Star Ratings technical notes. Updated March 21, 2019. Accessed January 19, 2020.
- Vanderholm T, Renner HM, Stolpe SF, Adams AJ. An innovative approach to improving the proposed CMS Star Rating “Statin Use in Persons with Diabetes.” J Manag Care Spec Pharm. 2018;24:1126-1129.
- Tsuyuki RT, Rosenthal M, Pearson GJ. A randomized trial of a community-based approach to dyslipidemia management: pharmacist prescribing to achieve cholesterol targets (RxACT study). Can Pharm J (Ott). 2016;149:283-292.
- Al Hamarneh YN, Hemmelgarn BR, Hassan I, et al. The effectiveness of pharmacist interventions on cardiovascular risk in adult patients with type 2 diabetes: the multicentre randomized controlled RxEACH trial. Can J Diabetes. 2017;41:580-586.
- Cai H, Dai H, Hu Y, et al. Pharmacist care and the management of coronary heart disease: a systematic review of randomized controlled trials. BMC Health Serv Res. 2013;13:461.
- Lowrie R, Lloyd SM, McConnachie A, Morrison J. A cluster randomised controlled trial of a pharmacist-led collaborative intervention to improve statin prescribing and attainment of cholesterol targets in primary care. PLoS One. 2014;9:e113370.
- Ho PM, Lambert-Kerzner A, Carey EP, et al. Multifaceted intervention to improve medication adherence and secondary prevention measures after acute coronary syndrome hospital discharge: a randomized clinical trial. JAMA Intern Med. 2014;174:186-193.
- Centers for Medicare & Medicaid Services. 2019 Medicare-Medicaid Plan Performance Data Technical Notes. Updated April 2019. www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-Coordination-Office/FinancialAlignmentInitiative/Downloads/MMPPerformanceDataTechNotes.pdf. Accessed May 20, 2021.
- Rodriguez F, Maron DJ, Knowles JW, et al. Association of statin adherence with mortality in patients with atherosclerotic cardiovascular disease. JAMA Cardiol. 2019;4:206-213.