X hits on this document

479 views

0 shares

0 downloads

0 comments

49 / 98

Results from the HIE indicated the average expenditure for prescription drugs in a plan

without cost sharing ($82) was nearly twice as high as in the plan with a 95 percent coinsurance

rate ($46) (Newhouse et al. 1993). In large part, this difference was driven by the number of

prescriptions

filled

rather

than

a

difference

in

the

average

cost

per

prescription.16

The

researchers concluded that the elasticity of demand for prescription drugs was –0.17, similar to

the elasticity of demand for health care in general.

More recently, Smith (1993) used cross-sectional claims data to study the effect of an

increase in copayments on the number of prescriptions purchased. He estimated a price elasticity

of –0.10 with respect to the number of prescriptions filled, but found no effect on total

expenditures for prescription drugs.

Motheral and Fairman (2001) examined the effects on prescription drug use among

employees enrolled in a PPO plan, moving from a two-tier cost sharing plan for prescription

drugs to a three-tier plan. Using a difference-in-difference research design, they estimated a

copayment elasticity of –0.21 with respect to use and –0.24 with respect to expenditures. Also

using a difference-in-difference approach, Gibson et al. (2005) compared employees facing an

increase in the copayment for prescription drugs with employees for whom copayments had not

changed, and estimated a much lower price elasticity (–0.04) with respect to use. They estimated

a –0.03 cross-price elasticity between generic and brand drugs, confirming that generic drugs

were substituted for brand-name drugs—a finding consistent with other studies as well (Goldman

et al. 2004). Finally, they found that the copayment effect on prescription drugs utilization

diminished after the initial reponse, suggesting that the long-run elasticity of demand is lower

than the short-run elasticity of demand.

16 In general, this result mirrors findings with respect to the impact of cost sharing on other service types—that is, cost sharing affects the probability of using any care, but not the cost of care once accessed (Ringel 2002).

39

Document info
Document views479
Page views480
Page last viewedSat Dec 10 11:35:04 UTC 2016
Pages98
Paragraphs2978
Words29067

Comments