Branded drug spending as an investment that is eventually paid off

TURN THE PAGE

In last week’s newsletter, I brought up how both health economists and drug developers use models to inform their resource allocation decisions, but the factors considered in these models differ, as does the time horizon.

In cost-effectiveness analyses used by health economists, a potential “reward” for a drug may be defined as a per-person annual price that would generate a cost-effectiveness estimate in line with some threshold. In NPV models used by drug developers, the potential “reward” is defined not only by price, but also by volume and margins.

The time horizon for a cost-effectiveness analysis is often the remaining lifetime of a patient who starts the drug at launch, whereas a time horizon across the product’s lifecycle is more typical for an NPV model to account for the initial development phase all the way through patent expiry.

Last week, I suggested that more economic models should estimate the total (accounting for the number of people potentially treated over the product lifecycle) long-term benefit of a drug over its lifecycle.

I want to expand on what I mean by this and why. Dr. Lou Garrison and colleagues did this very thing in a recent publication to examine the economic value generated by the direct-acting antivirals for hepatitis C. They used an economic model to estimate per-person expected health gains and health system cost savings and then multiplied these per-person estimates by utilization data (for years 2015 to 2019) to estimate the total benefit of these direct-acting antivirals for people who received treatment from 2015 to 2019.  

They found that the use of direct-acting antivirals from 2015 to 2019 results in a total benefit of approximately $580 billion due to patient health gains and health system cost savings (the direct-acting antiviral acquisition costs were not included in this step). Subsequently, they compared this economic value generated from the direct-acting antivirals to revenue numbers reported by the manufacturers. From 2015 to 2019, the 4 manufacturers of direct-acting antivirals captured 6.5% of the total economic value created by direct-acting antivirals—leaving society to capture 93.5% of the economic value.

For those who like math, the next four paragraphs are for you (for those who don’t, feel free to skip). From their economic model, they estimated that the use of a direct-acting antiviral would result in 4.4 quality-adjusted life years (QALYs) and $104,400 of cost savings (not considering the cost of the treatment) when used by a person who would not have received the previous standard of care (pegylated interferon and ribavirin). Using these traditional economic model outputs, you can calculate the per-person benefit by multiplying the QALYs gained by a threshold (they used $114,000 per QALY) and then adding in the cost savings. So the per-person lifetime benefit for a person who would not have received the previous standard of care was $606,000 [(4.4*$114,000)+$104,400].

For the proportion of the eligible population who would have received the previous standard of care, they estimated that use of a direct-acting antiviral would result in 1.7 QALYs and $41,500 of cost savings (not considering the cost of the treatment) per person treated. So the per-person lifetime benefit for a person who would have received the previous standard of care was $235,300 [(1.7*$114,000)+$41,500].

The per-person lifetime benefit estimates use the traditional outputs of an economic model. The novel part comes in here. The authors then estimated the total benefit by multiplying these per-person estimates by utilization data. Utilization data suggested 1,080,000 patients received direct-acting antivirals in that five-year period. It was assumed that 880,200 (81.5%) of them would not have received the previous standard of care and 199,800 (18.5%) of them would have received the previous standard of care. So they multiplied each number by the respective per-person benefits to calculate the total benefit: (880,200*$606,000)+(199,800*$235,300). This equated to a total economic value of $580 billion.

For those yelling “but you didn’t consider the cost of the drug!”, they compared the total economic value of $580 billion to the total revenue numbers reported by the manufacturers. Over that five-year period, the 4 manufacturers reported a total of $37.4 billion in revenues. Dividing $37.4 billion by $580 billion gets you to the 6.5% of the economic value that was captured by the manufacturers.

This 6.5% of the economic value captured by the manufacturers is likely an overestimate (as the authors acknowledge). The full product lifecycle wasn’t evaluated, but rather the analysis focused only on utilization during a 5-year period (of which all 5 years were free from generic competition). The proportion of the value captured by the manufacturers will continue to go down (and conversely the proportion of the value captured by society will continue to go up) as branded competition and future generic competition continue to reduce the cost of the direct-acting antivirals. Additionally, “value” in their economic model was defined based only on health system cost savings and patient health gains. If other outcomes such as patient productivity and caregiver impact were incorporated, the percentage of the total value captured by the manufacturer would be even lower (and conversely the percentage of the value captured by society would be even higher).  

Typical summary measures from a cost-effectiveness analysis are an incremental cost-effectiveness ratio (hypothetical example not linked to direct acting antivirals: $77,000 per quality-adjusted life year gained) or a value-based price (hypothetical example not linked to direct acting antivirals: an annual price of $42,000 would meet commonly cited cost-effectiveness thresholds). These summary measures are confusing to interpret, but even more so, they put the emphasis on the money, making us lose sight of the impact of healthcare innovations on patients and society. I argue that calculating the total long-term societal benefit over the product’s lifecycle is easier to comprehend and better informs the value that is captured by the manufacturer and how much is captured by us as a society.

The optimal percentage of the total value that the manufacturer should capture versus the percentage that society should capture is a topic of frequent debate in health economic circles, and current estimates vary widely. Although an “optimal” percentage is not yet known, presenting economic value estimates like this can put a spotlight on the value that society captures for its investment in a branded drug.

NOTHING ELSE MATTERS

In the sentence above, I intentionally called the money that we spend on branded drugs an investment. Too often we think of spending on branded drugs as an expense. I will raise my hand for being guilty of that. Rather than being short-term budgetary thinkers, we need to have a long-term investor mindset to see drugs as what they are—an investment that can generate returns in costs and health and are eventually paid off.

Branded drugs are unique in that they have a patent protection period. During that patent period, we are investing our pooled resources to pay for that innovation and to incentivize future ones. Fortunately for us as a society, that patent protection only lasts for a fixed period of time, after which generic/biosimilar competition can enter and prices need to substantially fall (and policies should be in place to make sure of it).

What if I told you that there was an opportunity to pay $34,630 and you would get something worth $537,037 in return? That’s what happened with the direct-acting antivirals for hepatitis C. The US paid $37.4 billion and got $580 billion in return.
For those who love math: $34,630 is the 5-year direct-acting antiviral revenue estimate ($37.4 billion) divided by the number of people who used direct-acting antivirals over the 5-year period (1,080,000 people), and $537,037 is the total economic value estimate ($580 billion) divided by the number of people who used direct-acting antivirals over the 5-year period (1,080,000 people).

Why don’t we have the same sense of pride or feeling like we got a massive return on our investment when it comes to branded drugs? I get that we aren’t all personally receiving that benefit, but we also aren’t individually paying that full amount. Rather, we pay into an insurance pool. I also get that not all the returns are direct dollars and some are health outcomes, but isn’t health important to all of us? And before you come at me, I get that direct-acting antivirals are revolutionary treatments and not all drugs will generate the same magnitude of benefits, but the fundamental principles hold for other branded drugs as well.

I recently watched this video of biotech executives who are all working to create new medicines for autoimmune diseases. They talked about how the portion of our insurance premiums that are used to pay for branded drugs are “creating the incentives” for ongoing innovation. At the end of the video, they encourage us to see the insurance deduction on our paycheck not as a dollar amount taken out of our pay “but the promise of scientific progress.” They hope that you will “see the brighter, healthier future that you are helping to build.”

For those of you who are still reading, you may be questioning the philosophy of thinking of drug spending as an investment. I get it. It’s not how I was first wired to think about it either, but one of the goals of this newsletter is to challenge conventional ways of thinking and start a conversation with more people.
What would help us see branded drugs as an investment rather than an expense? How can we appreciate returns in societal health gains even if they aren’t in direct dollars to us personally? As a health system and society, how do we move away from short-term budgetary thinking to value long-term returns?