- outcomes assessment (health care)
- professional role
- quality of health care
- reimbursement incentive
- risk sharing
The changing landscape of healthcare payment models is providing new opportunities for pharmacists. Traditional models were focused on paying for a product or a service. Evolving models are focused on paying for positive healthcare outcomes. One of the emerging payment models, the performance-based risk-sharing arrangement (PBRSA), focuses on medication-related outcomes.1These arrangements, historically between a pharmaceutical manufacturer and a healthcare payer, describe how payment for a drug product will be determined on the basis of prespecified healthcare outcomes. Since medications are a cornerstone of the treatment plan for many disease states, pharmacists, as medication experts, provide valuable contributions to achieving positive healthcare outcomes. There are several opportunities for pharmacists’ roles to evolve under PBRSAs, and this evolution is a natural extension of the value that pharmacists have already demonstrated in the healthcare system.
Practice areas such as antimicrobial stewardship and managed care pharmacy have generated evidence of pharmacists improving healthcare outcomes.2 In antimicrobial stewardship, pharmacists optimize the utilization of antibiotics, improve the health outcomes of patients, and ultimately reduce costs outside the immediate supervision of physicians. Antimicrobial stewardship programs, primarily in inpatient settings that include pharmacists, have been shown to produce annual savings of $200,000–$900,000, largely due to the reduction of antimicrobial drug use. There has been a similar impact of pharmacists within managed care. Payers leverage the expertise of pharmacists to aid in the development and execution of policies to promote the cost-effective use of medications.3 In managed care, pharmacists’ roles and responsibilities include drug dispensing, patient safety, clinical program development, drug-use guidelines, participation in pharmacy and therapeutics committee reviews, communications with clinicians and patients, drug benefit design, business management, and cost management. When pharmacists have developed guidelines requiring that patients meet Food and Drug Administration (FDA)–approved criteria for a given medication in order to improve the risk:benefit ratio of medication use, the literature has demonstrated associated reductions in health plan expenditures.4 In addition, there is evidence that when pharmacists conduct medication therapy management (MTM) for patients with hypertension or hyperlipidemia, there is improvement in Healthcare Effectiveness Data and Information Set measures as well as a savings of $12 for every dollar spent on MTM.4,5 The emergence of PBRSAs presents another opportunity for pharmacists to provide value to the healthcare system. Involvement in PBRSAs will require the application of pharmacists’ skills and knowledge in a new arena. This article describes PBRSAs and discusses how pharmacists participate in these new payment models.
What are PBRSAs? PBRSAs are known by many names, including risk-sharing agreements, patient access schemes, and managed entry schemes; the terms coverage with evidence and pay-for-performance are also used.1,6 In the context of a medication-focused PBRSA, the performance of a drug, which is determined by measures that vary from arrangement to arrangement but include cost and clinical outcomes, is tracked in a defined patient population over a period of time. The level of payment or reimbursement is based on whether or not the intended outcomes are achieved. PBRSAs are a natural response when payers have difficulty justifying coverage of high-cost medications and question whether outcomes observed in clinical trials will be achieved in practice.
In 2011 the International Society of Pharmacoeconomics and Outcomes Research (ISPOR), a global scientific and educational organization striving to optimize the use of health economics and outcomes research in healthcare decision-making, implemented a task force to define and characterize PBRSAs.1 The ISPOR task force’s definition of PBRSAs includes the core concept of tying the level of payment or reimbursement for a product or service to the real-world health and cost outcomes achieved within a specific patient population over a specified period of time. This idea is relatively radical: No longer is the payer taking on all the risk for adopting new expensive medications; rather, risk is to be shared with the manufacturer. In the context of PBRSAs, if anticipated clinical outcomes are not achieved in practice, manufacturers could be exposed to reduced reimbursement. The ISPOR task force outlined the infrastructure needed in order for these arrangements to work, also noting that accurate data collection is a necessity.
When to utilize PBRSAs. There needs to be a sufficient level of uncertainty about the real-world performance of a product relative to outcomes reported in clinical trials in order for manufacturers and payers to consider the use of a PBRSA.1 For example, questions from payers about the effectiveness and safety of new, higher-cost therapies relative to comparable low-cost standard therapies could delay the adoption of the new treatment strategy. The presence of this circumstance would incentivize manufacturers to engage in PBRSAs to encourage a payer to adopt their products. Given the amount of resources required for PBRSA implementation, not all therapies are appropriate targets. Cost–benefit assessments are necessary to determine whether meeting the information and infrastructure requirements of PBRSA implementation is practical and whether the PBRSA will add sufficient value for both payer and manufacturer. There are additional costs that payers must consider when deciding whether to engage in a PBRSA, such as the opportunity cost and the costs of delaying access to medication. Since PBRSAs are likely to be in place for discrete time periods, evaluating the costs associated with exit scenarios is also important, as altering medical practices can result in unexpected errors and negative consequences.7 These exit scenarios must consider the cost of educating clinicians on changes in practices.
Not all PBRSAs are similar. The ISPOR task force report made clear distinctions between what are and what are not PBRSAs.1 Arrangements by which medication discounts are driven by tying pricing to dispensing volume or market share are not PBRSAs, as they do not incorporate clinical outcomes. As payers and manufacturers begin initial discussions to engage in a PBRSA, the ultimate goal of the arrangement will dictate the use of a specific PBRSA subtype: coverage with evidence development (CED) or performance-linked reimbursement. The main question to address in determining the subtype is whether the clinical information to be developed will focus on population-level outcomes or individual-level outcomes.
In the event that population-level outcomes are the primary driver of payer uncertainty, a CED approach will be the primary framework of a PBRSA.1 Within a CED program, prospective data collection is performed within a research framework to track outcomes and, ultimately, to inform future decisions about coverage. Future coverage decisions could be prespecified and trigger automatic changes in reimbursement (i.e., decisions could involve setting minimum outcome targets above or below which changes in reimbursement are triggered) or they could be leveraged as evidence to inform future negotiations regarding reimbursement.1,8 The Centers for Medicare and Medicaid Services (CMS) began implementing CED policies in the early 2000s.8 CED policies are a subtype of PBRSAs focusing on high-cost pharmaceutical agents with rapidly increasing numbers of indications, such as oncological agents, in an effort to generate evidence for future coverage decision-making. CED was utilized by CMS as a response to ensure the proper on- and off-label use of colorectal cancer therapies such irinotecan, oxaliplatin, cetuximab, and bevacizumab.
If the focus of a PBRSA is on tracking individual-level outcomes to determine the real-world cost-effectiveness of a medication, performance-linked reimbursement will be the primary framework.1 The outcomes tracked to measure performance can be surrogate or clinical outcomes. For example, within diabetes mellitus, an individual-level outcome could be a glycosylated hemoglobin (HbA1c) target or the development or progression of microvascular complications. Performance-linked reimbursement PBRSAs can be written to focus on a specified outcome (e.g., an HbA1c value of <8% at 6 months) or ongoing outcome achievement (e.g., an HbA1c value of <8% at all specified time points). Although population-level benchmarks can be compiled, performance on the individual level must still be measured.1 Unlike CED, performance-linked reimbursement PBRSAs always link reimbursement to a clinical outcome.
Uptake of PBRSAs. Compared with some nationalized European health systems, the U.S. healthcare system has been slower to implement PBRSAs. Beyond CMS, the few existing examples of PBRSAs in the United States have typically been restricted to managed care and specialty pharmacies.1,9 A notable agreement in 2012 between a manufacturer and a pharmacy benefit manager (PBM) included the multiple sclerosis medication interferon-β1a.10,11 The manufacturer provided a rebate to the PBM if the overall cost to the PBM of enrollee treatment with interferon-β1a was greater than the costs of alternative therapies or if patients remained adherent to the drug at a predetermined level. Payers and manufacturers have described optimism for the potential value and future use of PBRSAs.9 However, current stagnation of the growth of PBRSAs in the United States is largely blamed on fragmented healthcare systems and the increased effort, resources, and infrastructure required to implement the necessary record-keeping systems.
In light of the Affordable Care Act, pay-for-performance is being incentivized.12 Programs and initiatives that have begun as a result include accountable care organizations, hospital value-based purchasing programs, the Medicare Physician Quality Reporting System, and a Five-Star Quality Rating scale for Medicare health plans. As a result, health systems in the United States are beginning to develop the required record-keeping systems in response to pay-for-performance reimbursement.9,13 PBRSAs have largely been focused on therapeutic areas with high drug costs such as oncology and management of multiple sclerosis. However, recent FDA-approved drug products, priced at unprecedented levels for primary care therapies, have garnered the scrutiny of payers.11 Some examples include new direct-acting antivirals for hepatitis C treatment and the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors for hypercholesterolemia. While few PBRSAs are currently in place, there are commitments by the manufacturer of sacubitril–valsartan, which is used in patients with heart failure, to work with payers and health systems to develop such arrangements.11,14 The manufacturer hopes to encourage increased utilization of the medication at a fast pace, and payers may be more willing to include coverage for this therapy with the assurance of lower prices if outcomes do not meet expectations.
Beyond having reservations about the resources required to initiate PBRSAs, decision-makers have voiced concerns related to having personnel who are trained and able to mindfully create and implement PBRSAs at given institutions.1,9 The key elements of the ISPOR-recommended PBRSA framework to which pharmacists can contribute are design, implementation, provision of care, and evaluation (Figure 1).1 Given pharmacists’ track record in optimizing therapeutic decision-making, involvement in those areas is a natural extension of pharmacists’ responsibilities. Through such involvement, pharmacists can aid in the development of PBRSAs and ensure accurate relevant data collection to address the uncertainty about the effectiveness and safety of a given product within a payer organization or health system.
Sacubitril–valsartan as an example. In the PARADIGM-HF study, the novel combination therapy sacubitril–valsartan was found to be more efficacious than enalapril in reducing the occurrence of the composite primary outcome of death from cardiovascular causes and hospitalization in symptomatic patients with heart failure with a reduced ejection fraction (HFrEF) receiving standard-of-care therapies15; the absolute risk reduction associated with sacubitril–valsartan treatment was 4.7% over 27 months (number needed to treat = 21).
Compared with enalapril use, sacubitril–valsartan therapy was associated with lower rates of serum creatinine elevation and hyperkalemia after a 4-week run-in period.15 Symptomatic hypotension was more common with sacubitril–valsartan than with enalapril; however, there was no between-group difference in rates of discontinuation of treatment due to this adverse effect.
Neprilysin inhibition with sacubitril simultaneously enhances the beneficial physiological effects of natriuretic peptides, while valsartan inhibits the detrimental effects of overactivation of the renin–angiotensin–aldosterone system.16 Sacubitril–valsartan is approved by FDA for use in place of an angiotensin-converting enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB) alongside other indicated HFrEF therapies.
As of April 2017, sacubitril–valsartan is a branded product (Entresto, Novartis) with an average wholesale price of $485.33 per month of treatment at the target dosage (sacubitril 97 mg and valsartan 103 mg orally twice daily); in comparison, a comparable dosage of the ACE inhibitor lisinopril costs $1.98 per month.17 With approximately 50% of patients with HFrEF dying within 5 years of diagnosis, the estimated cost of lifelong therapy with sacubitril–valsartan exceeds $22,000 per patient.18 By current estimates, 2.2 million Americans are eligible for therapy.19
Disparity in the pricing of sacubitril–valsartan and ACE inhibitor or ARB medications, the uncertain translation of the compelling efficacy results of the PARADIGM-HF trial into real-world practice, concerns about unrealized safety issues with increased use of the medication in diverse populations, and the current pay-for-performance measures focusing on heart failure reimbursement rates have produced an ideal climate to allow for the likely growth of PBRSAs focused on sacubitril–valsartan and have prompted the manufacturer to seek PBRSAs.11,14 Besides implementation of required information technology infrastructure, the success of these arrangements is dependent on a trained and knowledgeable work force, and pharmacists are a logical choice to shepherd PBRSA success.
Design and implementation of PBRSAs. Patients with heart failure often have multiple comorbid chronic and acute disease processes.20 Pharmacists’ involvement in contributing to positive outcomes through the management and monitoring of patients with heart failure has been established.21–23 Using the conceptual framework for PBRSAs outlined by the ISPOR task force,1 pharmacists are poised with the knowledge and skills to assist in the development of PBRSA processes to ensure safe and effective use of sacubitril–valsartan (Figure 1). Pharmacists in nondirect patient care roles, along with those involved in direct patient care, will apply population-based principles to define PBRSA criteria and protocols and determine the financial components of risk-sharing arrangements. Those in direct patient care roles will ensure optimal management of the patient and collect data to allow for determination of criteria for participation in the PBRSA and assessment of outcomes. Specific processes used by pharmacists in developing criteria for the appropriate use of sacubitril–valsartan include processes for addressing
Indication for use (e.g., setting of patient selection criteria based on symptom severity and prior treatment strategies),
Dosage adjustment (e.g., development of a dosing protocol based on PARADIGM-HF trial data),
Monitoring to assess safety and effectiveness of therapy (e.g., tracking urgent care or emergency department visits or heart failure hospitalizations, monitoring for syncope, tracking serum creatinine and potassium levels), and
Conversion back to ACE inhibitor or ARB therapy should sacubitril–valsartan therapy not be tolerated.
For successful implementation of these processes, pharmacists need to train healthcare providers on their responsibilities. Healthcare providers need to understand the criteria for appropriate use and the requirements for data collection.
Provision of care within PBRSAs. Although clinicians in a number of health professions have the ability to monitor the safety and effectiveness of sacubitril–valsartan use, there is one clinical service that pharmacy takes ownership of and performs frequently: MTM.24 Many of the services traditionally conducted through MTM are required for data collection for PBRSAs. MTM services include 5 core elements: medication therapy review (MTR), development of a personal medication record, creation of a medication-related action plan (MAP), intervention or referral (or both), and documentation and follow-up. These elements could be targeted to assess medication adherence and health literacy and develop action plans when adverse effects or exacerbations of heart failure are experienced. Incorporating these responsibilities into data collection to meet PBRSA documentation requirements would allow the accurate and complete tracking of data necessary for PBRSAs that target sacubitril–valsartan. MTM can facilitate and supplement the accurate tracking of information related to sacubitril–valsartan use that is required for the provision of care and evaluation within PBRSAs, as illustrated in the following HFrEF scenarios.
Scenario 1. A patient whose regimen of sacubitril–valsartan is adjusted upward to the target dosage is hospitalized 1 week later with syncope and hyperkalemia. The pharmacist determines that the patient had been taking valsartan as a single agent and did not stop taking it when sacubitril–valsartan was initiated. The medication record is updated to describe the events, the pharmacist counsels the patient to discontinue valsartan use, and follow-up examination of prescription claims reveals no further use of valsartan.
Scenario 2. A patient thought to be receiving the target dosage of sacubitril–valsartan for the previous month is hospitalized for heart failure and dies suddenly on hospital day 2. MTR conducted on admission by the pharmacist reveals that the patient was taking sacubitril–valsartan once rather than twice daily along with metoprolol succinate. Without MTR, the patient’s death might be classified as involving a failure of sacubitril–valsartan therapy.
Scenario 3. A patient with HFrEF comes to a health-system clinic with signs and symptoms of fluid overload and is subsequently admitted to the hospital. The pharmacist determines that the patient was not adhering to recommended sodium intake limits. Without review of the MAP, the admission could have been incorrectly classified as involving a failure of sacubitril–valsartan therapy.
Scenario 4. A patient arrives at the emergency department after having been hospitalized 3 weeks previously because of complications of HFrEF. He has been using guideline-directed therapy that includes target dosages of sacubitril–valsartan and metoprolol succinate; additional therapies include furosemide and spironolactone. The pharmacist determines that the patient adhered to all medication and dietary recommendations. It is determined that the 30-day readmission could be attributed in part to a failure of sacubitril–valsartan therapy.
Evaluation of PBRSAs. A 5-year pharmacy forecast released by the American Society of Health-System Pharmacists (ASHP) Research and Education Foundation in 2015 highlighted data technology integration of electronic health records (EHRs).25 Based on the input of a 159-member forecast panel (FP) of pharmacy experts, 74% of respondents predicted that about half of all health systems will implement EHRs that incorporate data from out-of-network providers. The FP also predicted that in at least 25% of health systems, patient care pharmacists will have umbrella responsibilities, encompassing both inpatients and outpatients, for pursuing the best outcomes from drug therapy.26 It is reasonable to believe that some duties currently performed by pharmacists will be applied to the collection of data for PBRSAs. These data are required to ensure that the medication is being evaluated within the contractual arrangement of the PBRSA.
Conclusion. The evolving healthcare marketplace has created opportunities for pharmacists within new payment models. Medication-related PBRSAs are a natural response to market changes reflecting stakeholders’ desires to distribute risk, especially in the current environment of increasing drug costs. Pharmacists can ensure the accuracy and consistency of the information necessary for the execution of medication-related PBRSAs. Critical PBRSA elements that can benefit from pharmacist involvement are development of policy, implementation, providing and documenting care, and evaluating data. Interoperable data communication and support from administration are necessary to ensure the successful involvement of pharmacists.
Kelly Calabrese, B.S., is acknowledged for her editorial input.
An audio interview that supplements the information in this article is available on AJHP’s website at www.ajhpvoices.org.
This article will appear in the July 1, 2017, issue of AJHP.
Dr. Cooke has consulted for and has received research grants from Novartis. The other authors have declared no potential conflicts of interest.
- Copyright © 2017 by the American Society of Health-System Pharmacists, Inc. All rights reserved.