ICER’s Cost Model Is Not Only Wrong It’s Also Dangerous
There they go again.
In the midst of the race for an effective COVID-19 treatment the Institute for Clinical and Economic Review (ICER) has performed an incomplete analysis of remdesivir in order to produce a cost estimate that is, by definition, precisely wrong. Remdesivir, produced by Gilead Sciences Inc., is an experimental antiviral medication that did not work as hoped to treat Hepatitis C, but is showing promise as a treatment for viruses such as SARS and COVID-19.
If ICER’s analysis were just an academic exercise, then the consequences would be minimal. But, there are short-term and long-term consequences from their analysis. In the short-term, ICER’s attempt to value remdesivir will make it harder to overcome the current pandemic. In the long-term, these exercises will diminish our ability to respond to the next one.
Let’s start with the basics.
According to the report last updated on May 1, 2020, ICER ran two models: a cost-recovery model that claims the price should be $10 for a 10-day course and $5 for a 5-day course; and a cost-effectiveness model that “suggests a price of approximately $4,500 per treatment course, whether that course is 10 or 5 days.” While surrounded by jargon and impressive sounding methodologies, both of these methodologies are without scientific justification because they depend on a litany of assumptions that cannot yet be known.
ICER notes that they “set the costs of research and development to zero” in order to calculate their suggested cost-recovery price for remdesivir. The justification for this bizarre assumption is that remdesivir already exists and underwent clinical studies “as part of a suite of agents for potential use in chronic Hepatitis C.”
Setting aside this “sunk cost” theory they are espousing, this assumption is clearly wrong. Gilead has incurred all sorts of R&D costs to determine whether this drug is effective against COVID-19. These include both the direct expenditures to run accelerated clinical trials and the large organizational costs to redirect resources from other projects in order to participate in this “all hands-on deck” response to the global pandemic. Ignoring these large costs mean that ICER’s cost recovery model estimate is wrong by definition.
As for their “sunk cost” theory, that is wrong too. The entire drug approval system is predicated on innovative companies receiving a patent for their treatments in order to recoup their capital costs once they are approved therapies. This opportunity ensures that the incentive for future innovations continues. Remdesivir has not yet been approved by the FDA (in the U.S. it has only received an Emergency Use Authorization, EUA, from the FDA effective May 1, 2020). Denying Gilead Sciences this opportunity violates the social construct that guides the drug approval process in the U.S.
If ICER’s advice on “sunk costs” is followed, it creates a terrible precedent that dis-incents any company from exploring whether a drug in clinical trials, or a failed drug, can address pressing global health problems.
Then there is the problem of societal value. ICER claims that, based on the cost-effectiveness model, the price should be $4,500 per treatment course, but this price clearly ignores the non-health benefits an effective treatment creates.
COVID-19 has brought the U.S. economy to a standstill, and the size of these economic losses is staggering. Starting with the impact on jobs, the April Jobs report showed that 20.5 million Americans have lost their jobs due to the pandemic, driving the unemployment rate to 14.7% — Great Depression unemployment levels. There are now tens of millions of families who have lost their source of income because of this pandemic and would highly value an effective treatment that could help them get back to work.
Beyond the lost jobs, the pandemic has drastically reduced the size of the U.S. economy. While we still don’t know the extent, Morgan Stanley is forecasting a 38% drop in economic activity (GDP) in the second quarter of 2020. Using Morgan Stanley’s estimated decline, and based on the $21.5 trillion U.S. economy (GDP as of 2020 Q1), the U.S. economy could have lost $8.2 trillion (on an annualized basis) in the second quarter – or around $25,000 per person. Morgan Stanley may be wrong. But even if the decline were only 10%, the economic losses would still be around $6,500 per person.
These staggeringly large numbers demonstrate the consequences from ICER’s assumption to ignore these broader social benefits from an effective treatment. Including the dollar value of lost jobs and lost economic output into their cost estimate would significantly increase the drug’s value. And, these only include the economic impacts. The negative costs from the social distancing rules and lost experiences matters too, further increasing the value of the drug.
Of course, all of these numbers are speculative, because we are still discovering the effectiveness of remdesivir. So far, the results have been positive. Perhaps, as scientists and doctors learn more, they will discover that remdesivir is more effective than current studies indicate, or it can be combined with other drugs to increase its effectiveness. Perhaps it is less effective than we hope. The bottom line is that it is premature to be discussing the cost of a medicine when its value is still being discovered.
This leads to the real danger of the ICER report. This report throws out an arbitrary number that appears to have been rigorously calculated, but in reality is nothing more than a guess. Based on the errors noted above, an awfully bad guess at that.
Should policymakers listen to ICER, they will be punishing the private companies and individuals who are stepping up to help solve this pandemic. The basic logic of incentives tells us that fewer people will do what is right if you punish them for doing the right thing. Essentially, this is what ICER’s cost model does.