sharing by changing their patterns of practice—for example, by prescribing fewer but more
intense treatments. As the patient’s agent, providers may either induce demand (to maximize
revenue) or constrain demand (in response to payment incentives such as capitation). Studies
that estimate price elasticity rarely account for provider behavior, or they include it in only the
most general terms—for example, controlling for care provided through an HMO. Failure to
control adequately for provider behavior may produce elasticity estimates that, while roughly
correct in the aggregate, may offer little insight about consumer responses to price changes in
different care environments.
B. PRESCRIPTION DRUGS
Many private and public health insurance plans have increased cost-sharing for prescription
drugs for the purpose of reducing demand overall and encouraging the substitution of lower-cost
drugs when possible. The introduction of greater cost sharing has offered researchers various
opportunities in the form of natural experiments to study the impact of greater cost sharing on the
use of prescription drugs. Nevertheless, programs that have developed more innovative cost
sharing designs—such as that in the Medicare Part D program—have had very little research
evidence to rely on.
Range of Estimates
Many insurance plans use “tiers” of cost sharing to control the use of pharmaceuticals. In
general, tiered cost sharing imposes higher copayments for drugs that have generic or less costly
equivalents. A number of recent studies have investigated the effects of tiered cost sharing on
consumer behavior. All of these studies suggest that the overall demand for prescription drugs is
price-inelastic, with estimates ranging from –0.10 to –0.60, depending on the data sources used
g., claims data from one employer or multiple employers) and the types of drugs considered
g., any drug or a specific class of drugs).