Value-Based Designs Boost Prescription Usage as Much as 15 Percent

A new modeling tool that measures the financial impact of value-based health plan designs may persuade more employers to implement them, its creators hope.

    The tool, which was developed by researchers at Harvard Medical School and the University of Michigan using several large, national employers as test subjects, initially was used to determine the medical efficacy of implementing value-based health plan designs, which emphasize high-value medication and services by lowering or eliminating co-payments to encourage plan members to use them.

    But Hewitt Associates has found a way to adapt the tool to help employers make implementation of a value-based health plan cost-neutral by projecting the additional plan costs if members avail themselves of the newly discounted drugs and services so that employers can, in turn, raise the co-payments for other less-valued drugs and services.

    Although the value-based insurance design concept is nearly a decade old, the early adopters—such as Pitney Bowes Corp.; Marriott International Inc. and Juno, Florida-based utility company FPL Group Inc.—pretty much did so based on “faith” rather than science, benefit experts acknowledge.

    Although these companies publicly encouraged more employers to follow their lead, “the later adopters are more skeptical and hesitant,” said Craig Dolezal, a senior health care consultant for Hewitt in Atlanta.

    “The universal question that employers ask is if there always is an increased cost,” he said. “This model takes prescription drug cost and utilization data at the most detailed level, calculates and organizes the data to show compliance … then imports the data into a clinical actuarial model” that employers can use “to calculate the cost and utilization impact on their plan and how to spread those costs.”

    For example, “if an employer wants to add incentives for diabetes and heart medications, but it doesn’t have additional money to spend, the tool will help determine how much to increase costs for other therapeutic classes and non-chronic condition-treating drugs to offset the additional cost,” Dolezal said.

    “Value-based designs are attractive because they are designed in a way that removes financial barriers to care and encourages employees to seek and receive the essential care they need to manage their health,” said Melissa Miller, director of employee benefits and services at FPL, which was among the employers involved in the development of the Hewitt tool.

    “But the biggest question for many employers is how to design these programs in the most cost-effective way,” she said in an August statement announcing Hewitt’s release of the tool.

    Before now, most studies have measured the negative impact on utilization of raising drug co-payments, but little research has examined lowering co-payments, said Clive Riddle, founder and president of Managed Care On-Line Inc., an online medical and health benefit technology vendor based in Modesto, California, who was not involved in the development of the Hewitt tool.

    Since “cost shifting to the patient has been the big mantra this decade,” Riddle said, “usually studies look at the opposite: If they raise co-payments, what impact will it have on utilization?”

    Riddle cited a study of two employer-sponsored drug plans published in 2003 that found a significant number of employees stopped taking certain medications—rather than switching to generics—when their employers moved to a prescription drug plan that charged significantly more for brand-name drugs.

    Moreover, any estimates of potential savings of implementing value-based insurance designs “have historically been not research- or tool-based,” he said. “It’s been more quick-and-dirty, comparing projected expenses against actual.”

    “They look at the prior year’s drug spend and try to project, but they’re always off,” Hewitt’s Dolezal agreed. “Quantifying of the investment is a really big step for employers.”

    “For a number of employers, this type of tool would give them a much better comfort level. These kind of tools are required to get the next layer of employers to move” into value-based health plan design, Riddle said.

    Using a tool to reallocate health care expenses also “allows companies to meet cost targets without sacrificing the quality of care their workers receive,” said Michael Chernew, professor of health care policy at Harvard Medical School in Boston, a collaborator on the tool’s development.

    “The other question is to figure out how much savings would be produced by reducing adverse events that might occur if a person suffering from a chronic condition does not take their drugs,” he said.

    “Value-based design initiatives recognize that it is imperative that as employers continue to control health care costs, they also have the responsibility, and the vested interest, in doing so in a manner that does not jeopardize employee health. The principles behind value-based insurance design, and the tool developed by Hewitt Associates, enable employers to meet these two important goals,” Chernew said in an August statement announcing the new tool.

    “What the tool is designed to do is to allow a firm to meet whatever financial target they want to meet in a way that incorporates value-based insurance design as opposed to across-the-board clinically insensitive ways,” Chernew said in an interview.

    “Before value-based design, they would just raise co-pays across the board to meet a financial target. But this doesn’t recognize the differences in value for different services,” he said.

    Chernew and Dr. A. Mark Fendrick, director of the Ann Arbor, Michigan-based Center for Value-Based Insurance Design at the University of Michigan, published the conceptual framework for value-based plan design nearly a decade ago, and founded the center, which was created to promote, implement and evaluate value-based designs.

    New York-based ActiveHealth Management has been using a similar tool, also developed in conjunction with Chernew and Fendrick, in a three-year pilot program involving Marriott, said Dr. Steve Rosenberg, senior vice president of outcomes research at the disease management and clinical decision support firm. ActiveHealth also has technology that identifies patients in particular need of interventions who are not getting them for one reason or another, he said.

    “We use it in two ways: If a company or a health insurer is interested in implementing the value-based insurance design and they want to know the estimate of the cost or savings, we can run this model on their data to produce an estimate,” Rosenberg said. “For customers we already have, we use the model to report to them every six months on the savings their program is achieving.”

    While not divulging individually identifiable health information, the tool “measures how many people receive notices” telling them about the program and “how many comply after getting the notice,” he said. “We also measure the people already taking the drugs to see how many continue taking them.”

    In general, “we find there’s a 10 percent to 15 percent improvement in long-term compliance when co-pay is reduced or eliminated,” Rosenberg said.

    In addition, “what we find routinely is that the cost for medication for the employer always increases, the non-drug costs for the employer are reduced by the same amount or a little more, so it’s basically a wash,” he added.

    In the Marriott study, which was reported at the March annual meeting of the Washington-based National Business Group on Health, generic co-pays were eliminated and brand-name drug co-payments were halved for the types of medication used to treat diabetes, asthma and heart disease.

    Researchers are still studying the experiment’s outcomes, but “as we implement this technology, we are finding these strategies are creating a positive difference in trend,” Jill Berger, Marriott’s vice president of health and welfare, said at last spring’s National Business Group on Health meeting.

Workforce Management, October 22, 2007, p. 23 — Subscribe Now!