Characteristics of high–drug-cost beneficiaries of public drug plans in 9 Canadian provinces: a cross-sectional analysis

Like all drug utilization analyses, the article by Tadrous et al. provides only one side of the pharmaceutical equation. It provides no information about the outcomes of the drug use because the National Drug Utilization Information System (NPDUIS) cannot be linked with clinical data. Canada has a major lack of large-scale clinical data (i.e. electronic medical records, not diagnoses in administrative data) that can be linked with drug prescription data.

Even the data on the use side of the equation is lacking. As the authors have done, assumptions can be made that sofosbuvir/ledipasvir was prescribed for hepatitis C and insulin glargine for diabetes, but several of the other drugs have multiple uses. Furthermore, these inferences do not provide any insight into the severity of the diagnosis, which is critical to be able to evaluate whether the drug use was appropriate.

Tadrous et al. also do not acknowledge that the NPDUIS only records prescriptions dispensed and does not account for (a) differences in drugs listed in provincial formularies - for example, Prince Edward Island does not list sofosbuvir, (b) copayments, deductibles and premiums, which vary by province - copayments and deductibles are means-tested in all provinces, but out-of-pocket payments vary widely from $100 in Ontario to several thousand dollars in some western provinces, and (c) provincial differences in clinical criteria that must be met before a patient can be considered for access to many of the high-cost medicines. These differences impact access, especially for high-cost medicines.

Based on a limited, narrow snapshot of data, the authors raise two concerns: the use of costly medicines and the use of large numbers of medications. The suggestion seems to be that neither are appropriate. However, without clinical evidence and an understanding of the differences outlined above, an assessment of appropriateness cannot be made.

An analysis of surgeries performed in Canada would find that some, such as brain surgeries and transplants, are much more expensive than others for more simple conditions. No one would suggest that further work should explore targeted interventions to address surgery costs, especially if information about the reason for the surgery and the severity of the diagnosis was not considered because it was unavailable.

Patients who need expensive surgery do so because, without it, their health will decline and they may die. Provincial governments and Canadians support such treatment. Patients who need expensive medications for hepatitis C, HIV/AIDS, cancer or rare disorders do so because, without them, their health will be negatively impacted and they may die. Why should there be a question about the need for costly medicines and not one about the need for expensive surgeries?

In the commercial world, insurance is designed to provide assistance for major events that impact lives by spreading premiums across many individuals, most of whom receive no financial benefit from their premium. For example, home insurance pays out for significant, financially-damaging events such as fire or water damage or theft, which means relatively few people receive a lot of benefit, a larger number receive a modest benefit and the majority of the insured receive no benefit, other than knowing that if they need coverage, it is there.

This situation also occurs in the health system, relatively few patients receive a lot of treatment, more receive some and many receive little. The same should apply to public and private drug insurance. However, numerous articles, like that of Tadrous et al., suggest that public coverage for the relatively few patients who require costly medications is somehow inappropriate but patients receiving coverage for low-cost medications is appropriate. This is not how insurance is intended to work.

Drug insurance should be available to assist patients who would otherwise be unable to pay for the costly drugs they need for conditions such as hepatitis C, HIV/AIDS, cancer and rare disorders, like paroxysmal nocturnal hemoglobinuria or atypical hemolytic uremic syndrome, that without treatment are debilitating (leading to significant costs to other parts of the health system) and fatal. It should not simply be coverage for inexpensive drugs.

As the federal government considers the development of national pharmacare, a system based on stringent price regulations and a parsimonious formulary will significantly reduce treatment options for more Canadians, potentially leading to poorer health outcomes that will satisfy no one.

Conflict of Interest:

In the past three years, I have received consultant fees from 3Sixty Public Affairs Inc and Fasken, research and publication fees from Advocacy Solutions, the Canadian Health Policy Institute, the Fraser Institute, Medicines New Zealand, Merck Sharp & Dohme (New Zealand) Ltd, RAREi (a collaboration of innovative pharmaceutical companies focused on the development of medicines for rare disorders) and Ward Health, and publication processing expenses from Canadian PKU and Allied Disorders Inc and Shire Pharma Canada ULC.