John Graham's concerns about the effect of reference-based pricing on drug expenditures in British Columbia present an opportunity to further clarify the situation described in our commentary.1
Reference-based pricing was expected to produce the most savings for elderly patients (those covered by PharmaCare Plan A), because they were the primary users of reference drug classes: antihypertensives, nitrates and NSAIDs. According to official PharmaCare statistics,2 there was minimal growth (0% to 2.6% per patient annually) in Plan A expenditures between 1995 and 1997, the time of active expansion of reference-based pricing, but this growth increased to 8% to 10% later.
Successful drug cost containment does not necessarily lead to a reduction in global drug expenditures, particularly if the elderly population is increasing rapidly, as is the case in British Columbia. However, it should significantly slow the increase in per capita expenditures in the target population. Furthermore, after the introduction of BC's PharmaNet network in late 1995, reimbursements have been provided automatically rather than being based on submissions of claims. This change resulted in a surge in reimbursements to patients under 65 years of age who had previously been unaware that they could receive coverage after reaching a certain level of expenditure. Global budget comparisons across the country are therefore unhelpful.
Successful drug policies such as reference-based pricing should not lead to lower prescribing rates but to a shift toward more cost-effective alternatives where available and toward newer breakthrough drugs where needed. Our research provides evidence that this was achieved.3,4 Other investigators have independently come to the same conclusion. Morgan5 showed on an aggregate level that changes in drug mix during the expansion of reference-based pricing led to substantial savings for PharmaCare while overall utilization was unchanged.
Unfortunately, Graham has misinterpreted the results of our study published in the New England Journal of Medicine.3 Because only 14% of those using ACE inhibitors switched to lower-priced ACE inhibitors, the primary comparison was between those who switched drugs and those who did not. As we discussed at length, it is difficult using claims data to fully adjust for confounding by patient health status when this status is a predictor for both future hospital admissions and switching to a lower-priced ACE inhibitor (because of more frequent physician encounters if health status is poor). Follow-up beyond the 2-month period Graham mentions is therefore more meaningful. The rate ratio for changes in hospital admissions was 1.19 (95% confidence interval [CI] 0.99–1.42) for 2 months and even lower, with a tighter CI, for 10 months (1.03, 95% CI 0.92–1.14).
Sebastian Schneeweiss Colin Dormuth Division of Pharmacoepidemiology and Pharmacoeconomics Brigham and Women's Hospital Harvard Medical School Boston, Mass.
Competing interests: None declared.
References
- 1.Schneeweiss S, Maclure M, Dormuth C, Avorn J. Pharmaceutical cost containment with reference-based pricing: time for refinements. CMAJ 2002; 167(11):1250-1. [PMC free article] [PubMed]
- 2.BC Ministry of Health Services. Pharmacare trends 2002. Victoria: The Ministry; 2002. p. 14, 24.
- 3.Schneeweiss S, Walker AM, Glynn RJ, Maclure M, Dormuth C, Soumerai SB. Outcomes of reference pricing for angiotensin-converting-enzyme inhibitors. N Engl J Med 2002;346:822-9. [DOI] [PubMed]
- 4.Schneeweiss S, Soumerai SB, Glynn RJ, Maclure M, Dormuth C, Walker AM. Impact of reference-based pricing for angiotensin-converting enzyme inhibitors on drug utilization. CMAJ 2002; 166(6):737-45. [PMC free article] [PubMed]
- 5.Morgan SG. Quantifying components of drug expenditure inflation: the British Columbia seniors' drug benefit plan. Health Serv Res 2002;37: 1243-66. [DOI] [PMC free article] [PubMed]