Want to Be in a Health Plan? Think Like One

It happens in health plans all over the country: Pharmacy and Therapeutics Committees meet to consider the placement of a new drug on the formulary. A drug company representative has provided the dossier, including a cost-effectiveness (CE) model that meets the requirements laid out by the Academy of Managed Care Pharmacy (AMCP) Format for Formulary Submissions. It seems like a slam dunk. The model demonstrates that the drug is cost-effective; while it costs 50 percent more per pill and has a rare-but-catastrophic side effect, it results in a 50 percent increase in quality-adjusted life years (QALYs) compared to another treatment.

Jill Van Den Bos

Then the committee starts asking questions:

• How will the older-than-average age of our population affect outcomes?

• How does this intervention stack up against other available treatments?

• QALYs are too subjective. What about the reimbursement, utilization, and morbid-outcome assumptions made in modeling these data?

• What are the expected impacts on pharmacy, hospital, and physician budgets?

Clearly, these questions go beyond the generic cost-effectiveness results typically provided by manufacturers. Such questions include requests for information tailored to a health plan, with overt specification as to the greater impacts of formulary decisions on the health plan as a whole as well as its members and specific demonstration of value compared with competitors.

While CE model results are worthwhile, especially in coming to a yes-or-no coverage decision—e.g., "the $45,000 per QALY gained passes our test for inclusion as a preferred drug"—they don't help with monitoring of experience or health plan pricing. To do this, the information provided would need to show elements of a health plan budget, as well as how utilization and per-member per-month (PMPM) claim costs may be impacted by the addition of the new drug. This is most useful for pricing and monitoring of experience compared with expectation, both important financial functions within a health plan.

The Model Cost Model

Consequently, pharmaceutical man-ufacturers should consider the importance of providing information that catches the attention of those outside the P&T Committee, particularly the actuarial department. Pharmacoactuarial analysis produces economic models that can help companies meet AMCP guidelines and demonstrate a treatment's value in a way that formulary committees can understand and use.

Demonstrating Value

An actuary specializes in the mathematics of risk, especially as it relates to insurance calculations such as premiums, dividends, reserves, and insurance and annuity rates. A primary function of health plan actuaries is to ensure that the premium charged for coverage will be adequate for the health plan to cover claims and other expenses. The financial health of the health plan relies on the actuarial department's calculations.

This analysis is based on modeling that incorporates outcomes and dollars, and utilizes actuarial inputs or techniques. Actuaries also incorporate variables such as trends, dynamic-population analysis, and adjustments for other factors like plan-benefit design (e.g., covered services and copayments) and internal financial structure (e.g., financial incentives for providers). Actuaries tend to create outcome measures in terms of PMPM cost rates, rather than composite measures like cost-effectiveness ratios. This is how health plans price.

A PMPM cost represents the total expected dollars spent for each enrolled person per month on all claims, possibly including health plan expenses. This value is obtained by pooling the total cost experience of a large covered population and averaging this amount over the total exposure period of that population (in months). By contrast, a cost-effectiveness ratio represents the cost per some specific outcome, such as cost per cases of disease averted or cost per life year added. The first measure covers all health plan services aggregated, while the second is specific to the intervention under study. Therefore, the first metric is useful for broader issues surrounding health plan costs (e.g., pricing, monitoring experience of the population), while the second is useful for determining the value of a specific intervention alone.

Consider this example: There's a drug with frequent side effects that are minor but result in other scrips being written to manage them. A new, more expensive drug comes along; it has few complications, but one rare side effect always results in costly surgery. A CE model could show that the increased effectiveness of the new treatment resulted in an increase of X number of quality-adjusted life years per dollar over the older treatment. An actuarial budget-impact model, however, could dynamically model cost and utilization impacts of the treatments on all health plan costs and produce a PMPM cost comparison of the two treatments. If it could be shown that the new drug not only improved quality of life but also reduced costs in unexpected areas (e.g., ER visits), this would make for a compelling value proposition.

Actuarial analysis considers the health plan budget as a whole, and by nature focuses on the probable total-dollar impact of decisions. The business-focused perspective of actuaries gives them a unique advantage in producing studies that speak to the needs of budget-conscious health plans.

The Actuarial Approach

The actuary's main contribution to the economic research toolkit is the actuarial budget-impact model, also called a cost model. Because the techniques used are drawn from the experience of healthcare actuaries—who are employed most commonly by insurers and health plans—such models are uniquely suited to application in the formulary submission process.

The AMCP definition of budget impact model falls short of what an actuary would do. It reads:

"These models are used to evaluate the budget impact of pharmaceutical expenditures as a consequence of new product introduction. ... By strict definition, budget impact models are not used to establish the value of new technology because they do not include the impact of drugs on clinical outcomes, nonpharmacy resource use, and adverse effects."

A well-constructed budget-impact model does include impacts outside of the pharmacy budget, and it quantifies them in greater detail than is found in a traditional CE analysis.

Risks, Rewards

Perhaps the most difficult barrier to overcome in implementing a pharmacoactuarial approach is a "side effect" of its own utility: The actuarial budget-impact model is so good at comparing different treatment options that manufacturers might consider it difficult to conclusively demonstrate the value of their own products. Understandably, they may be tempted to reject the services of actuaries.

In the long run, however, there is little benefit to not implementing a pharmacoactuarial approach. Health plans will continue to engage in aggressive management of the pharmacy budget, and will adopt whatever techniques are shown to most accurately predict the true ratio of cost to value for new treatments. And for many new regimens, especially those with a high up-front cost, the actuarial budget-impact model is in fact the best way to conclusively demonstrate value, by showing positive impacts across the organization.

An even more effective strategy from the drugmaker's point of view is to incorporate pharmacoactuarial thinking into the product development process, examining the potential for a new offering to reduce overall health budget impacts in innovative ways. At this point, the actuary becomes a partner in designing the most effective economic information for the health plans.

As the pharmaceutical industry becomes even more competitive, and as health plans continue to demand conclusive demonstrations of the value of new offerings, drug companies will need all the proof they can get in order to succeed.

Jill Van Den Bos is a consultant with the health practice at Milliman. She can be reached at jill.van.den.bos@milliman.com