Pharmaceutical treatments and medical devices often have varying effectiveness depending on the indication for which they’re used: in oncology, for instance, response to a treatment varies with the type of tumor and stage of disease. The advent and proliferation of precision medicine in which biomarkers — whether genomic, proteomic, or structural — identify patients likely to receive greater treatment benefits only increase the range of variability in the effectiveness of the same product.
Yet manufacturers traditionally charge the same price for all indications. Recently, there have been calls for “indication-based” pricing systems, in which manufacturers are paid more when treatments are used for indications for which they have higher value (“high-value indications”) and less for indications for which they confer less benefit (“low-value indications”).1,2 Supporters hope that such a system will reduce prices for low-value indications but that prices for high-value indications will not increase.1 This expectation arises from a belief that manufacturers currently set uniform prices according to the value generated for high-benefit indications and somehow get patients who receive lower value to pay the same price.
If that were true, it would mean that manufacturers have convinced insurers and patients to consistently pay prices exceeding their products’ value. But simple economics makes it clear that relative to uniform pricing, indication-based pricing results in higher prices for patients who benefit the most, higher utilization by patients who benefit least, higher overall spending, and higher manufacturer profits.
Pharmaceutical prices result from negotiations between manufacturers and payers (insurers), both of which want to maximize profits. To do so, payers must consider how their customers will respond to incremental premium increases that result from passing along higher pharmaceutical prices. If patients value access to a drug, they will be willing to pay the higher premium; if they don’t, insurers must find ways to limit how often their enrollees use a high-priced pharmaceutical (in order to keep spending, and therefore premiums, from increasing) or face losing customers. For example, payers often implement utilization-management tools such as prior authorization and step therapy to limit access to a drug to only those patients who value the drug the most.3 This strategy leaves manufacturers facing the classic monopolist’s trade-off: setting higher prices reduces the quantity sold, but setting lower prices reduces the revenue obtained from the patients who are willing to pay the most.
To illustrate this process, the dollar value of a product was calculated (by multiplying some measure of the treatment’s clinical benefit, such as the expected increase in the length of life, by the price patients or payers are willing and able to pay for that benefit) for three different indications, the size of the patient populations affected by each indication, and the prices charged for the product. Effects of Uniform Pricing versus Indication-Based Pricing.. In this framework, lower prices result in greater access to the product by populations for whom it has lower value.
In Panel A, the upper graph represents a uniform-pricing context in which patients with indication A receive the most benefit and those with indication C receive the least. The population with indication C is large, and the value of treatment to this group is close to the value for indication B. As a result, the manufacturer’s profit-maximizing price allows all patients to obtain the drug. At this price, the manufacturer earns profits represented by the green area.
But the firm faces a trade-off. By setting the price in this way, the manufacturer forgoes profits that could be earned by charging higher prices to patients with indications A and B. These forgone profits, represented by the blue areas, are captured by these patients as consumer surplus — the value difference between the most consumers are willing to pay and what they actually pay.
The lower graph in Panel A shows a different scenario, in which the product’s valuation for patients with indication C is very low. In this case, it’s a better trade-off for the manufacturer to set a high price, at which it knows the payer will allow only patients with indications A and B to obtain the drug. The manufacturer accepts the loss of sales to patients with indication C in exchange for higher profits earned from patients with indications A and B. Hepatitis C cures such as sofosbuvir fall into this category: at current prices, these drugs are unavailable to many insured patients who either have viral genotypes on which the treatment is less likely to be effective, have relatively little disease progression, or have no coexisting conditions.4
Comparing these graphs, we see that when the valuation of the product for indication C is relatively low, manufacturers set a higher uniform price, the payer curtails sales to patients with indication C (orange area), and patients with indications A and B obtain less consumer surplus than they did in the first scenario.
What would happen if the manufacturer used indication-based pricing — setting a price that more closely matches the product’s value to each customer? This is a practice that economists call price discrimination, and its effects are well understood. In the most extreme version, the manufacturer extracts the most money each patient is willing to pay, leaving no consumer surplus. This scenario can be seen in the graphs in Panel B, which reveal three key points: regardless of the proportion of patients with each indication, all patients can obtain the treatment; patients who receive the most value from the drug pay the highest price; and pharmaceutical profits are higher because more drugs are sold and marginal costs are exceptionally low.
For the purposes of clarity, we’ve simplified some realities of insurance. For example, Medicare and many states mandate coverage for all cancer drugs. Such mandates make it harder, but not impossible, for insurers to affect consumption — for example, they can still use utilization-management techniques, and thus our analysis still applies.3
To provide a sense of the potential magnitude of the effects of indication-based price discrimination, we use data from Peter Bach’s pricing proposal for cetuximab (Erbitux), an epidermal growth factor receptor (EGFR) modulator.1 When used as first-line treatment for recurrent or metastatic squamous-cell carcinoma of the head and neck, cetuximab is associated with a median survival gain of 0.23 years, and it costs $10,000 per month. The patients in question would be those with indication C in the upper graph in Panel A. Payers assume that as long as patients value access to a therapy more than $10,000, they will buy insurance that provides access to it.
When cetuximab is used to treat locally advanced squamous-cell carcinoma, by contrast, it offers a median survival gain of 1.64 years. Under uniform pricing, patients with high-value indications (indication A) also pay $10,000 per month. But if we apply the same value for survival improvements in these patients as that implied by the $10,000 price paid for low-value indications, manufacturers implementing indication-based pricing could charge $220,000 per month for their treatments (Panel B). Absent indication-based pricing, the manufacturer could not set such a high price without having payers reduce access for patients with low-value indications — the trade-off would not be worth the lost profits.
So what would indication-based pricing accomplish? For drugs currently priced so high that they’re unavailable for some indications, it expands access. Drug manufacturers would now be willing to set low prices for low-value indications, since it wouldn’t jeopardize their profits on high-value indications. But the same access-expanding pricing flexibility also allows manufacturers to increase prices for high-value indications. Currently, some treatments are priced low enough to be accessible for a wide range of indications, and it is there that we should expect the biggest price increases.
The growth of precision medicine creates an ideal setting for manufacturers to deploy predictive biomarkers as tools of price discrimination to capture more of the overall value created by their products.5 Payers who don’t understand the underlying economics may go along under the banner of “value-based pricing.” But the trade-offs involved should be carefully considered.
1. Bach PB. Indication-specific pricing for cancer drugs. JAMA 2014;312:1629-1630
CrossRef | Web of Science | Medline
2. Pearson SD, Dreitlein B, Henshall C. Indication-specific pricing of pharmaceuticals in the United States health care system: a report from the 2015 ICER Membership Policy Summit. Boston: Institute for Clinical and Economic Review, March 2016 (https://icer-review.org/wp-content/uploads/2015/03/Final-Report-2015-ICER-Policy-Summit-on-Indication-specific-Pricing-March-2016_revised-icons-002.pdf).
3. Rockoff JD. How Pfizer set the cost of its new drug at $9,850 a month. Wall Street Journal. December 9, 2015 (https://www.wsj.com/articles/the-art-of-setting-a-drug-price-1449628081).
4. Dickson V. As insurers limit access to hep C drugs, patients and doctors bristle. Modern Healthcare2015 (http://www.modernhealthcare.com/article/20150520/NEWS/150519897).
5. Stern AD, Alexander BM, Chandra A. How economics can shape precision medicines. Science 2017;355:1131-1133
CrossRef | Web of Science | Medline
This Perspective article originally appeared in The New England Journal of Medicine.