'ICER' strikes again - Scandal time?
This week, the self-proclaimed drug-price watchdog ICER issued a "Report on Unsupported Price Increases", picking a number of innovative medicines that, in the view of the authors, have seen hikes without sufficient substantiation. While the study made the headlines, we may want to pause for some skepticism and ask, how sufficient is the ICER analysis?
This opinion piece was first published on PharmaBoardRoom, here.
Let's agree on the fundamentals and state at the outset: The U.S. has a ridiculous problem with the cost of goods for healthcare. Compared to several other developed countries, we pay more for worse outcomes. A study published this week in JAMA found that roughly 20%- 25% of all U.S. health care spending is wasteful. A second trend, more similar to other countries, is that medical inflation is a massive concern. And, again unlike most other developed countries, our current insurance system has seen ordinary patients pay a high price for these excesses. Scott Gottlieb, while still FDA Commissioner, called the status-quo a Kabuki theater where the "sick are subsidizing the healthy". The situation is totally unacceptable, and if you get angry looking at the boom of personal insolvencies just due to medical cost, mi casa es tu casa, enough is enough. In this environment, the new ICER bombshell made the evening news. Once again, it alleges with "independent" credibility, do we see the drug industry costing us billions, a tragedy of the commons and result of abject market failure. A wonderful narrative for the 2020 campaign, uncritically picked up by trade and national media alike. Unfortunately, ICER's piece is flawed in so many fundamental aspects that it should concern us just how much credibility we continue to bestow on the group - however angry we are, and wherever we stand on drug pricing or pharmaceutical innovation.
To begin with, the basic premise of selecting 9 medicines with drug price increases to make the argument of unsubstantiated price hikes ignores fundamental logics in both the micro- and macroeconomy of pharmaceutical innovation. Individual companies must have the liberty to price successes to be able to sustain failures in the quest for future cures (price optimization cannot be a government-run decision unless the government also wishes to run global R&D programs). On the macro-level, patents purposefully protect market monopolies as part of a societal contract that sees generics enter the market at a fraction of the cost once brand exclusivity periods end. It can be argued that incumbents try to exploit this in several cases, but by and large, the societal deal has payed-off in the US: A low estimate puts the number of prescriptions that are today filled for generics at 8 out of 10. Curiously, Cialis, one of the 9 products in the ICER report, is available as a generic and highly accessible irrespective of the manufacturer's pricing. We can debate the system, but that is precisely how it has been designed to function.
Economics aside, when it comes to the report's actual analysis, the real problems begin. ICER simply miscalculates the price manufacturers actually realized from those products it selected for the public shaming. When that became obvious to researchers in at least a dozen cases yielding a net price above the (gross) list price, or WAC, the authors decided to simply remove reference of the list price for those products. But they continue with the flawed methodology nevertheless for all others as if the inability to properly calculate the net prices only pertained to the extreme case where the math went so far south that net (somehow) came out higher than gross. Consequently, at least one company openly disclosed to ICER that based on the developer's actual data, not the outside ICER math, net price changes for the assessed products were, in fact, negative. Meaning, the innovator made less money on the branded medicine than in the prior year. That is probably not a bad thing, you might say. Unfortunately, ICER chose to ignore addressing that important correction which would obviously invalidate its methodology here. Leaving a short "Thank you for this information" in the response section, the report goes on to cite the list price increase % which ICER had misestimated, not the decrease that occurred. Missing out on countervailing evidence breaks the neck of any social science research, reproducibility. When we openly disregard evidence we know to be true and replace it with our version of what reality feels like, then we basically find ourselves in the opinion section.
But do we know what happens to list prices? Well, it turns out that concerns around drug price inflation have produced helpful data points on just how much of the value is actually captured by pharma companies. Such research entirely contradicts the ICER narrative and continues to be ignored across all of ICER's work. Net price increases are independently estimated to be below medical inflation (data above from IQVIA). If you are still skeptical, (after all IQVIA, like this author, has clients among drug companies), let's turn to Express Scripts, one of the nation’s largest Pharmacy Benefits Managers and the main antagonist of the pharma industry in the drug price blame-game. The PBM's data cuts reveal that spending on medicines in commercial plans grew just 0.4% in 2018 net of rebates and discounts, the lowest in 25 years. Why does this matter? Well, if list prices climb and net prices stay below inflation, it is the Kabuki theater of reimbursement and rebating that is subsidized here. A win for PBMs perhaps, but neither patients nor drug manufacturers see any gains. And if pharma were to have applied modesty, what happened to the unjustified price hikes ICER claims to have uncovered? All of a sudden, the story would become a lot less egregious.
A report claiming "unsupported drug price increases" that also "cannot determine whether the increases were justified"
We learn that ICER's stated objective in this search was not to actually evaluate the evidence base, nor to establish what a value-based price for these products would be. It attempts to do just that in its other reports, but for this one, it seemingly didn't find the time or resources:
"It is important to note that ICER does not have the capacity to perform full economic analyses on the nine therapies evaluated in this report, nor would the time needed to develop full ICER reports (at least eight months) provide information in a useful timeframe for the public and policymakers. Therefore, this report is not intended to determine whether a price increase for a drug is fully justified by new clinical evidence ... "- Page ES2
What, one might ask, should we make of the analysis then? We learn that time is scarce for those who are intended to get the message, and they apparently don't require a "full" analysis to arrive at our conclusion. The goal, the report states, was to see whether "substantial new evidence existed that could justify (sic)" an increase. Confused? Not for no reason. If you parse those words, it is verbal gymnastics. If the report has the stated limitation of not being able to determine whether a price increase is justified, it cannot be titled 'Report on unsupported price increases". That is unless we are to infer that a price increase may have well been economically justified but that it doesn't mean it was supported by evidence. Is there another benevolent category of justified price increase that we are unaware of? There is not. 'Unsupported' price increases are not 'justified', is what ICER claims, without fully wanting to say out loud that this is a blatant contraction with its own statement. It either is or is not a limitation for the research, you either can or you cannot say it. If you are a bit mischevious or spend your days in social science research settings, you might wonder that if we are uncertain enough to state the limitation on our core finding in the introduction, wouldn't it be required to have secured the capacity to be able to find out for sure, or alternatively aim lower with our claim? That supposes that there still was a claim if we didn't grab for the headlines.
Let us assume that it doesn't matter and we accept the analysis at hand prima facie. What ICER went on to do, is what makes the entire concept most worrisome and indeed, it matters more. It asked manufacturers for the additional evidence they could produce on the drugs, manufacturers complied and send it over, only to see ICER selectively ignore dozens of studies and hundreds of literature references that fell into the ICER-chosen assessment window. "Outside of our scope" appears 92 times in the 125 pages of the document. There is no question that the scope determined the analysis, it always does per nature of the exercise, and that it was at best highly "restrictive", at worst entirely flawed to answer the supposed research question. ICER is very transparent on the fact that it discounted a lot of publications, it is not so clear what exactly qualifies discounting in the first place. For instance, many of the studies showed improvements in quality of life benefits and leveraged real-world evidence on the products ICER selected. Manufacturers, such as AbbVie, pointed out correctly, that ICER's own study protocol for the systematic review was selectively applied.
"AbbVie provided more than 200 scientific publications that support the value of HUMIRA and its safety and clinical effectiveness. Despite these being peer-reviewed publications – many of which have been presented at major medical congresses around the world – ICER determined that none of the evidence fully met the review process criteria. It is AbbVie’s position that in rejecting all of AbbVie’s submitted research from consideration, ICER has excluded high-quality evidence of added net health benefit of HUMIRA that would be appropriate to consider under the assessment protocol." - Page 75
Yes, Humira carries most of Abbvie's bottom-line, its price is very, very high. But many of us have long demanded industry to become more patient-centered, now that it does, we cannot just dismiss any evidence it brings to the table as propaganda and discount observational research that runs counter to our hypothesis perse as "low-quality". To be clear, there is no way of determining to what degree ICER has done this for all of the research it had received here. But its pronouncedly restrictive approach notably breaks from the FDA which now considers these evidence types as meaningful endpoints even for approval, and it breaks with payers who want to see real-world patient effectiveness, not just controlled trial data. ICER further validates the growing concern among patient associations who have long challenged the group on its lack of interest to properly incorporate patient perspectives. Those articles ICER dismissed as irrelevant were at the minimum peer-reviewed and published in scientific journals. Posters were presented at established conferences where part of the medical community discussed the research impact. Surely, innovators would try to garner excitement, some very likely therapeutically inconsequential, and we should take all of it with a grain of salt. But do a few authors at ICER have the unilateral ability to determine what counts for medical progress across such a variety of therapeutic areas?
Even when ICER produces its more complete analyses (the type it didn't have time for here), stakeholder concerns have surfaced. For instance, when the PBM CVS suggested to base a formulary on ICER's regular value assessments in 2018, the backlash from patients and doctors was so enormous that CVS didn't move forward with the plan (and has not to date) since it could not reconcile the stakeholders' views with the challenges in the methodology. Let us even argue that the ICER authors who made that important call on our behalf did have the almighty capacity, guided only by the benevolent interest to uncover the truth, driven by a sophisticated and objective approach that meets the task... If we assumed all of that, we might still want to know what mandate the group has owned to make these determinations. As much as a drug developer ought to be questioned for the obvious motive to increase its own products' market share, we cannot just state fairly obvious research flaws without addressing why ICER is doing this type of work in the first place. Who put them in front of us? Is there more than meets the eye?
In media and many academic circles, ICER orbits as the independent quasi-governmental agency in the service of helping us determine what responsible drug pricing would amount to. It has become the HTA America didn't have, using the cost-effectiveness toolkit that has kept prices low in places such as the UK. Whether or not you agree with the premise that a central body (public or not) should make coverage decisions for a population of 350 million, that one body should be allowed to dictate in a free-market economy just what price levels are justifiable... at least, everyone would concur that said body ought to be accountable and impartial, not carrying favor for any particular lobby of the healthcare industry. Sadly, in 13 years of its existence, ICER has not meaningfully addressed the cost problem of 85% of the healthcare dollar burdening Americans - that is all those costs that have nothing to do with pharmaceutical spending. It has certainly never addressed the insurance system that ruthlessly exposes patients to copays and co-insurances. For all its PR, ICER's health economists are seemingly unconcerned with applying their discipline to the areas that hurt patients the most. Its target group - stated openly as state and federal lawmakers in the current report - is to act on drug prices. That's it. In an America where the government warns its citizens that a three-day hospital stay can cost around $30,000 on average, ICERs watchdog myopia is at least a bit suspicious.
And, as so often in political D.C. when things are fairly obviously wrong, vested interests are not far from the action. ICER is overwhelmingly funded by "Arnold Ventures" run by the former hedge fund manager and Enron gas-billionaire John Arnold. Arnold Ventures is behind various other campaigns to lower prescription drug prices across the nation. Maybe most notably, it is also the main financier of the political action committee (PAC) "Patients for Affordable Drugs" which came to light in the 2018 Mid-term election by running barrages of TV attack ads against Congressmembers of the Democratic Party who would refuse to endorse Medicare for All. If this sounds a tad too conspiratorial, it is not. Neither ICER nor Arnold make any effort in hiding mission and agenda, as the WSJ recently profiled. In this week's report, ICER openly states that its new type of investigation, on price hikes, became possible only through the generous multi-year grant from Arnold. And the money was, so to speak, on target.
As a public, we need multiple voices in this debate, which includes strong views from outside of the pharmaceutical industry. But for the sake of intellectual honesty, we also have to be more aware who produces the views that do not originate from 'bad pharma'. Every time we 'obviously' discount a study by PhRMA as coming from the "industry lobby group", the 'opposite' research, paid for by the other side, is equally 'obviously' to be taken with caution. Knowing where the group comes from, we would then have to acknowledge that ICER is about as "independent" as Senator Bernie Sanders when it comes to the impartial arbitration of affordable drug prices.
As qualified as many of the researchers around Steve Pearson might be, ICER is not a public advocate on everyone's behalf, nor has it been mandated to provide academic analyses of objective status. It simply is one voice, paid for by one actor, like the voice from Abbvie quoted earlier. Some call it philanthropy, others see 'dark money'. Do affiliation and political funding make ICER authors automatically guilty of systemic bias? Possibly not, unless the work itself was affected by an obvious lack of objectivity and rigor. Unfortunately, this recent ICER report, judged on its merits, leaves little room but to conclude that it did matter who paid the bills here. ICER went data picking in support of a narrative. In the long-run, the group could well be providing one of many substantial perspectives we need. We may just want to stop elevating that one voice to be the 'independent' one in this important discussion.