The Impact of Reference Pricing of Cardiovascular Drugs on Health Care Costs and Health Outcomes: Evidence from British Columbia

Key Implications for Decision Makers

  • Drug plan managers (both public and private), health policy makers, pharmaceutical manufacturers, and clinicians and scientists should find this report useful.
  • Reference drug pricing programs work on the principle that if several drugs work equally well for a certain condition, the program will fully fund the drug that costs the least. Patients may choose the more expensive drug but the program will only reimburse people for the cost of the lowest cost reference drug(s). BC Pharmacare fully reimburses the cost of drugs for patients who need the higher cost drugs for medical reasons.
  • Reference pricing does save Pharmacare money. Savings depend on the drug. Reference pricing programs are most effective where there is a substantial price difference between drugs in the same class (in the ACE inhibitors category, the price difference was not that large, so the savings were negligible).
  • Many patients chose to pay out of pocket for the higher priced drugs. Patient expenses accounted for roughly one quarter of total Pharmacare savings.
  • Death rates did not increase when reference pricing was introduced for drugs used to treat cardiovascular or renal disorders (such as ACE inhibitors, calcium channel blockers and nitrates)
  • Reference pricing does not increase the rate of long-term care admissions for patients.
  • Patient visits to their doctors do increase when reference pricing is introduced, possibly to discuss whether to switch to a lower cost drug.
  • Low-income seniors who did not receive continuous exemptions from reference pricing used slightly fewer higher priced drugs and were more likely to switch to a fully subsidized drug, or reduce or discontinue drug use.
  • Because the authors are still uncertain about the effects of reference pricing on morbidity, it is unclear if the benefits of reference pricing outweigh the costs.

Executive Summary

Purpose of the project

Reference pricing has been adopted domestically and internationally as a way to reduce costs for drug subsidy and insurance programs. Reference pricing limits the reimbursement of a group of drugs with similar therapeutic effect but different active ingredients to a fixed "reference price". The setting of the reference price varies by jurisdiction but typically is based on an average of the lowest cost "reference standard" drugs within the group. Patients can buy the lower cost drug, or they have the option of paying the difference between the retail price and the reference price for partially subsidized drugs.

Considerable controversy was generated with the introduction of reference pricing by the British Columbia Ministry of Health Pharmacare program, the provincial government drug subsidy program for seniors and others. Critics of reference pricing contend that the partially subsidized and fully subsidized (reference standard) drugs are not therapeutically interchangeable, and patient health will therefore be compromised. Also, the use of physicians' and hospital-based health services may increase as a result, thus partially or wholly offsetting any potential cost savings from the policy. The reference pricing policy, however, allows selected patients to be exempted from the policy.

We investigated whether reference pricing has reduced Pharmacare drug expenditures without adversely affecting patient health status (either mortality or morbidity) or increasing expenditures on physicians' and hospital-based services. We also investigated the following distributive effects of the policy: How much did patients collectively pay for the partially subsidized "restricted" drugs? Did patients' access to the restricted drugs depend on their ability to pay?

Target audience

Drug plan managers, both public and private, health policy makers, consumers, pharmaceutical manufacturers, clinicians and scientists should find this report useful.

Recommendations and policy implications

From the perspective of the BC Ministry of Health, a comprehensive assessment of the reference pricing policy requires quantification of all of its costs and benefits. This study estimates the benefits but not all of the costs of the introduction of the reference pricing policy to senior Pharmacare beneficiaries.

On the benefits side, we estimate that the application of reference pricing to the nitrates, angiotensin-converting-enzyme (ACE) inhibitor and dihydropyridine calcium channel blockers (CCB) drugs was associated with a reduction in expenditures on its seniors drug plan in the order of $24 million as of May 1999. On the other side of the ledger are the additional expenditures on ambulatory physician consultations (likely under $1 million), and the costs of the additional revascularization procedures, and the cost of the additional sublingual nitroglycerin dispensed to nitrates users.

We have not yet estimated all of these costs because of uncertainty as to the validity of our models. A full assessment would also need to incorporate the costs of administration of the reference pricing program (the additional staff required to handle exemption requests, for example). Until these costs can be quantified, statements regarding the overall effects of reference pricing are premature.

Main findings

Drug costs and use
We estimate that reference pricing of the drugs we studied reduced Pharmacare expenditures on its seniors drug plan by $23.8 to $24.8 million (depending on the estimation method used) between October 1995 to May 1999. This translates into annualized savings to Pharmacare of approximately $7.7 million, or 3.6% of the $213.7 million that Pharmacare spent on drugs for seniors (not including dispensing fees) in 1997. Approximately 24% of Pharmacare's savings represent additional costs to seniors who elected to pay for the higher cost drugs. The drugs examined in this study were the nitrates, ACE inhibitors, and CCBs drug classes.

Mortality
Using conventional statistical methods, we found no evidence of an increase in rates of mortality associated with cardiovascular or renal disorders after reference pricing was applied to the nitrates, ACE inhibitors, and CCBs drug classes.

Morbidity and physician costs
Using conventional statistical methods, we found no evidence of an increase in the rates of longterm care admissions after reference pricing was applied to the nitrates, ACE inhibitor and CCBs drug classes.

The use and expenditures on ambulatory physician consultations increased after the implementation of the policy, likely due to the fact that many seniors needed to consult with their physicians to discuss treatment options (for example, to switch to a fully subsidized drug, or apply for an exemption), and to monitor the progress of those who switched drugs. The increased visits could have been due to changes in patient morbidity as well.

Sublingual nitroglycerin is used to control acute exacerbations of angina and was used as a marker for patient health status. Seniors using the nitrate drugs for angina that were no longer fully subsidized when reference pricing was introduced faced a higher likelihood of using sublingual nitroglycerin in the first 4 months of reference pricing. These subjects also had a higher likelihood of bypass surgery or other revascularization procedure in the longer run. No such effects of morbidity were observed for the application of reference pricing to ACE inhibitors or CCBs, although there was an increase in the rate of revascularizations among those taking ACE inhibitors, in the first 4 months after the policy only. The results of these morbidity models should be seen as tentative, until these results can be replicated using alternative estimation strategies. For this reason the costs of the revascularization and related services were not estimated.

Effect of income on access to the partially subsidized "restricted" drugs
Low income seniors who do not receive continuous exemptions from reference pricing use slightly fewer restricted drugs post policy and hence are more likely to switch to a fully subsidized drug, or reduce or discontinue drug use. Thus for those seniors who are not exempted from reference pricing, there is some evidence of income-based impediments to access for the restricted drugs, although the impediments do not appear to be particularly large.

Methodology

Using quasi-experimental methods, we focused on the effects of reference pricing on senior (65+ years) beneficiaries of the BC Pharmacare program. To investigate the effects of reference pricing on drug costs, we assembled monthly claims data, aggregated across all senior beneficiaries, on the drug groups targeted by reference pricing, and examined the pre-post reference pricing changes in prescribing volumes, Pharmacare reimbursement and patient expenditures for these drugs.

We used longitudinal patient-level data to estimate the effects of reference pricing on morbidity and non-drug health care expenditures. We also used patient-level data to estimate the effects of reference pricing on death due to cardiovascular disease and time to admission to a long-term care facility. We also modeled how senior's income status (an indicator of low income) affected both the probability of exemption from reference pricing, and, in the subsample of seniors who were not exempted, both the probability of any payment and the amount paid for restricted drugs.