Currently Viewing:
The American Journal of Managed Care May 2015
Comparison of Provider and Plan-Based Targeting Strategies for Disease Management
Ann M. Annis, MPH, RN; Jodi Summers Holtrop, PhD, MCHES; Min Tao, PhD; Hsiu-Ching Chang, PhD; and Zhehui Luo, PhD
Making Measurement Meaningful
Christine K. Cassel, MD, President and CEO, National Quality Forum
Care Fragmentation, Quality, and Costs Among Chronically Ill Patients
Brigham R. Frandsen, PhD; Karen E. Joynt, MD, MPH; James B. Rebitzer, PhD; and Ashish K. Jha, MD, MPH
Results From a National Survey on Chronic Care Management by Health Plans
Soeren Mattke, MD, DSc; Aparna Higgins, MA; and Robert Brook, MD, ScD
Association Between the Patient-Centered Medical Home and Healthcare Utilization
Rainu Kaushal, MD, MPH; Alison Edwards, MStat; and Lisa M. Kern, MD, MPH
Transforming Oncology Care: Payment and Delivery Reform for Person-Centered Care
Kavita Patel, MD, MS; Andrea Thoumi, MSc; Jeffrey Nadel, BA; John O’Shea, MD, MPA; and Mark McClellan, MD, PhD
True "Meaningful Use": Technology Meets Both Patient and Provider Needs
Heather Black, PhD; Rodalyn Gonzalez, BA; Chantel Priolo, MPH; Marilyn M. Schapira, MD, MPH; Seema S. Sonnad, PhD; C. William Hanson III, MD; Curtis P. Langlotz, MD, PhD; John T. Howell, MD; and Andre
Innovative Care Models for High-Cost Medicare Beneficiaries: Delivery System and Payment Reform to Accelerate Adoption
Karen Davis, PhD, APN; Christine Buttorff, PhD; Bruce Leff, MD; Quincy M. Samus, PhD; Sarah Szanton, PhD, APN; Jennifer L. Wolff, PhD; and Farhan Bandeali, MSPH
Annual Diabetic Eye Examinations in a Managed Care Medicaid Population
Elham Hatef, MD, MPH; Bruce G. Vanderver, MD, MPH; Peter Fagan, PhD; Michael Albert, MD; and Miriam Alexander, MD, MPH
Currently Reading
Systematic Review of Benefit Designs With Differential Cost Sharing for Prescription Drugs
Oluwatobi Awele Ogbechie, MD, MBA; and John Hsu, MD, MBA, MSCE
Medicaid-Insured and Uninsured Were More Likely to Have Diabetes Emergency/Urgent Admissions
Monica A. Fisher, PhD, DDS, MPH, MS; and Zhen-qiang Ma, MD, MPH, MS
Roles of Prices, Poverty, and Health in Medicare and Private Spending in Texas
Chapin White, PhD; Suthira Taychakhoonavudh, PhD; Rohan Parikh, MS; and Luisa Franzini, PhD
Measuring Patient-Centered Medical Home Access and Continuity in Clinics With Part-Time Clinicians
Ann-Marie Rosland, MD, MS; Sarah L. Krein, PhD, RN; Hyungjin Myra Kim, ScD; Clinton L. Greenstone, MD; Adam Tremblay, MD; David Ratz, MS; Darcy Saffar, MPH; and Eve A. Kerr, MD, MPH

Systematic Review of Benefit Designs With Differential Cost Sharing for Prescription Drugs

Oluwatobi Awele Ogbechie, MD, MBA; and John Hsu, MD, MBA, MSCE
A systematic review of insurance benefit designs with differential cost sharing for substitute prescription drugs.
ABSTRACT

Objectives: To evaluate the effects of health insurance benefit designs that introduced or increased the price difference between prescription drugs representing potential clinical substitutes.

Study Design: Systematic review of peer-reviewed articles.

Methods: Using English-language articles listed in PubMed between 1980 and 2012, we identified articles meeting our inclusion criteria and minimum methodological standards. We compared findings regarding the immediate patient response, total spending, and health outcomes after implementing the price change.

Results: Among the 31 articles identified, the mechanisms varied for creating the price differential between prescription drug substitutes, though they most frequently involved tiered formularies (19) or reference pricing (10). While nearly all studies (29 of 31) reported on patient responses to price changes, only 5 articles comprehensively assessed patient price responses, total spending, and health outcomes. Several studies found that some patients switched to cheaper drugs, but out-of-pocket spending increased on average, suggesting that other patients continued using the more expensive drug (ie, cost shifting to patients). Few studies examined the degree of heterogeneity in behavior responses, especially between patient cohorts for whom the substitute drugs had varying value. Some studies observed long-term effects, but most had limited post intervention observation periods.

Conclusions: Differential cost-sharing designs influence drug use behavior, but there is limited evidence on how these designs affect the overall value of received care. The existing literature provides limited guidance for policy makers or organizational leaders to design benefits. We offer suggestions for future studies to inform policy and practice.

Am J Manag Care. 2015;21(5):e338-e348
Take-Away Points

To our knowledge, this is the first systematic review of the literature on differential cost-sharing benefit designs that examines the effects of pricing explicit therapeutic substitutes differently to patients.
  • We found that these designs influence drug use behavior and healthcare spending, though the evidence is limited on whether they improve the value of care.
  • Few studies examine the effects of these designs on clinically relevant subgroups, such as the elderly or patients with multiple comorbidities.
  • Reforms, including those preceding and within the Patient Protection and Affordable Care Act, may increase the prevalence of these designs; therefore, understanding their effects is imperative.
Insurance plans and payers in the United States are increasingly introducing health insurance benefit designs with embedded financial incentives targeting clinical treatment decisions. These initiatives are part of a larger movement toward benefit designs that seek to change patients’ behaviors by aligning patient incentives with clinical goals through financial incentives. Value-based insurance design (V-BID) and consumer-directed health plans are examples of benefit designs that encourage patient involvement in decisions in part by manipulating their price of medical care to better reflect the value of that care.1,2 The overarching principle of such designs is the difference in out-of-pocket (OOP) price among various treatment substitutes, which creates the financial incentive that could influence the patient’s choice of treatment. Moreover, the value of a medical therapy to a patient depends on not only the perceived therapeutic effect, but also on the price of the therapy when compared with other alternative treatments. Hence, the price differential—or the difference in OOP price—of medical treatment substitutes contributes to the patient’s perceived value of the medical treatment and may affect patients’ medication use and adherence. Recent studies suggest that some types of incentive-based benefit designs could encourage higher-value treatment choices and promote lower growth of medical spending in the United States.3,4

Much of the cost-sharing literature has focused on healthcare utilization or adherence when the OOP price for most drugs in a drug class are concurrently increased or decreased.5 As many medical therapies have a degree of price elasticity, changes in price often affect use. However, crude applications of prices yield crude incentives that fail to reflect the value of the treatment options or the inherent clinical nuance necessary to determine value for real patients. Benefit design managers need to identify novel mechanisms to modify patient behavior and incentivize them to select the appropriate therapies for their condition. To date, insurance plans have applied incentive-based designs most often to prescription drugs, such as through mechanisms including tiered formularies, reference pricing, or free drugs for chronic diseases.1,2,6

Importantly, incentive-based designs are not synonymous with high cost sharing or tiered cost-sharing designs, as many manifestations of the latter designs frequently aim to discourage overuse to address the moral hazard associated with being insured or to shift costs from payers to patients. A common example of these cost-based, rather than value-based, designs is the placement of inhaled steroids into the higher cost-sharing tiers without lower-cost options. Inhaled steroids are critical drugs for the prevention of asthma exacerbations, but until recently, they have all been on patent with no existing generic versions that would have qualified for the lower cost-sharing tiers in many insurance plans.

Even when plans aim to encourage high-value treatments, the implementation is fraught with difficulty as real-world clinical decisions may involve substantial nuance. For example, the true clinical benefits associated with a drug could vary substantially across individuals depending on factors such as prior clinical history, genetic predisposition, or other current medications.7 Consequently, the potential of heterogeneity in treatment outcomes could mean that the individual value of each treatment and the substitution of seemingly equivalent treatments could differ significantly from simple population-level estimates of value. This potential heterogeneity reinforces the need to assess actual effects of these designs on health outcomes within relevant populations, especially if policy makers encourage, and organizations implement, thoughtful incentive-based designs.8,9

Recent studies examining the effects of these designs have mainly focused on the effects of changes of the absolute drug price, but fewer have examined the price differential between substitute drugs.5,10 This is important because the price differential creates the financial incentive for using one drug over another and influences treatment choices, rather than simply the decision to treat or the cessation of use. Furthermore, other reviews focus on evaluating the immediate or short-term effects of a specific form of benefit design, rather than the downstream effects.11,12

Although the literature on cost-sharing plans is extensive, the evidence to support incentive-based benefit designs is just emerging, as several studies suggest the potential to reduce drug spending with limited effects on drug adherence.6,11,13 Furthermore, the literature on incentive-based benefit designs has grown substantially in the last decade, but to date, there have been no systematic analyses to synthesize the new findings into a meaningful recommendation for administrators and policy makers.

In this study, we define incentive-based health insurance designs as designs that use differential cost sharing for prescription drugs with potential clinical substitutes, present a framework for their evaluation, report findings from a systematic review of the literature, and discuss suggestions for future research. Throughout this process, we take the perspective of a policy maker or organizational decision maker evaluating the evidence base to support decisions about creating new incentive-based programs.

METHODS

Definition

Incentive-based benefit designs use differential patient cost sharing to create meaningful incentives that alter drug use behavior. To be effective, these designs must also avoid unnecessary complexity and eschew unintended, clinically significant consequences.1,8 In their simplest form, these designs create a choice set of substitute drugs that are clinically similar, but have different cost-sharing amounts, which then creates an incentive to choose the lower cost-sharing option.14 When drug substitutes are more effective or less expensive than other alternatives, they have higher value, because the implied cost per unit of effectiveness is less than that for other choices.

Rationale for the Evaluation Framework

Major outcomes of interest to benefit managers, payers, providers, and policy makers include medical and pharmacy utilization, medical and pharmacy spending, and health outcomes.5 Because of the heterogeneity of clinical effects, including varying effects with age and clinical comorbidities, a thorough analysis requires information on the average overall effect and range of effects on relevant clinical subpopulations. Similarly, changes in context also could be relevant. For example, in the pharmacoepidemiologic literature, many studies differentiate between new and preexisting drug users, as the former represent a group making a new decision about treatment and use, whereas the latter often are more clinically heterogeneous in their treatment experiences, disease severity, and trajectory.5 With the latter, at least some of the patients could have attempted using multiple therapeutic options and failed to achieve satisfactory improvement.

Furthermore, information on both short- and longer-term effects are important since the effects of many prescription drugs could require years to manifest, as is the case with blood-glucose or blood-pressure therapy. As a parallel, many clinical trials differentiate between drugs that only achieve intermediate end points and those that improve overall health outcomes.

Data Selection

We identified quasi-experimental studies that used longitudinal changes in prices and evaluated the before-and-after effects of differential pricing of drug substitutes on drug use behavior, spending, and health outcomes. We explicitly excluded studies that did not include an evaluation of differential incentives on patient behavior (eg, excluded studies that increased all drug prices or made all drugs free). As stated earlier, multiple studies have explored the elasticity of medical treatments, but this study was geared toward examining patients’ behavior when patients had a clear choice set of treatments with differing prices. For the initial search, we queried the PubMed database for articles published in English from January 1, 1980, to December 31, 2012, using a combination of 2 sets of keywords for incentive-based designs and behavioral responses: 1) incentive-based design terms included a combination of terms for cost savings, cost sharing, benefit design, formularies, and reference pricing located in the title, abstract, or National Library of Medicine’s Medical Subject Headings (MeSH); and 2) behavioral-response keywords included MeSH terms for drug substitution, medication adherence, patient compliance, drug utilization, pharmaceutical services utilization, healthcare utilization, and outcome assessment. We excluded surveys, willingness-to-pay studies, behavior modeling, cross-sectional studies, retrospective cohort studies, case reports, and prior authorization interventions. We also excluded anonymous articles, news articles, editorials, letters, guidelines, interviews, meta-analyses, and other reviews.

After the initial search, we reviewed titles and abstracts to identify articles that analyzed or presented data examining the effects of an insurance benefit design change that introduced or increased differential pricing of substitute prescription drugs. We explicitly excluded articles that did not state the choice set of substitute drugs or the price differential between the drugs. We then reviewed the reference lists of the selected articles for additional studies that met the inclusion criteria. When a single differential pricing intervention resulted in multiple articles, we included all articles if the outcome measures differed across them. Finally, we queried experts in the field about potential articles.

Two researchers independently reviewed studies’ eligibility and inter-rater reliability was greater than 95% for article inclusion. We conducted the search according to the 2009 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for systematic reviews.15

Figure 1 displays the review process, starting from the 458 articles identified during the initial search to the 23 articles selected that met inclusion criteria, as well as the 8 additional articles from the reference list review. There were no additional studies meeting our inclusion criteria from the reference lists of the latter 8 articles. Additional details about the methods are available upon request.

Identifying Outcome Measures

 
Copyright AJMC 2006-2017 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
x
Welcome the the new and improved AJMC.com, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up
×

Sign In

Not a member? Sign up now!