Jesse Sussell, PhD

 To project the social value of transcatheter aortic valve replacement (TAVR) for inoperable patients with severe, symptomatic aortic stenosis (SSAS).

 This study used an economic model with parameters obtained from the literature and from US Census Bureau population projections.

 Our model estimated the economic value that will accrue to inoperable patients with SSAS and to device manufacturers as a result of TAVR utilization. We estimated individual patient value as the monetized gain in quality-adjusted life-years as estimated in the cost-effectiveness literature, net of device costs and cost offsets. We estimated manufacturer value by applying an assumed profit margin to revenue from device sales. We created population-level estimates by combining these individual-level estimates with age-stratified Census Bureau population projections and estimates of the incidence of AS. We assessed model uncertainty through the use of probabilistic sensitivity analyses.

 Between 2018 and 2028, approximately 465,000 inoperable Americans with SSAS will be treated with TAVR. These procedures will yield a cumulative social benefit of up to $48 billion, with roughly 80% of that benefit accruing to patients and 20% accruing to device manufacturers.

 Policy makers and payers should take this social value into account when considering decisions related to the care of inoperable patients with SSAS.

Considering the US population between 2018 and 2028, we estimated the annual and cumulative social value accruing to patients and to manufacturers as a result of uptake of transcatheter aortic valve replacement (TAVR) for the treatment of severe, symptomatic aortic stenosis (SSAS) in inoperable patients.

In 2018, approximately 39,000 inoperable patients with SSAS were treated with TAVR in the United States.

That number is projected to increase as the population ages and as TAVR uptake grows.

TAVR treatment of the inoperable population with SSAS will generate approximately $48 billion in social value between 2018 and 2028, with 80% of that value accruing to patients.

Aortic stenosis (AS) is a common disease, affecting 2% to 7% of the global population older than 65 years.

 Although the condition can be asymptomatic for years, when symptoms do appear, untreated patients with severe AS have high mortality (up to 50% within 2 years).

 Surgical aortic valve replacement (SAVR), in which the sternum is opened to replace the aortic valve, has been shown to improve survival and quality of life in patients with severe, symptomatic AS (SSAS).

 Despite data supporting the use of SAVR for the treatment of SSAS, at least 30% of patients do not undergo SAVR, because of either patient preference to avoid invasive surgery or ineligibility for surgery as a result of advanced age and associated comorbidities (the latter referred to as inoperable patients).

 Historically, this population of untreated patients was treated with medical management alone, resulting in 3-year mortality rates as high as 80%.

 Transcatheter aortic valve replacement (TAVR) is a less invasive method of AVR in which the aortic valve is replaced through either the femoral artery or a small incision in the chest. Clinical trial data have demonstrated that TAVR for inoperable patients extends median survival by approximately 19 months and significantly improves quality of life compared with medical therapy alone,

 thereby offering an alternative treatment for this patient population. Accordingly, TAVR was first approved for commercialization in Europe in 2007

 and by the US FDA in 2011 for the treatment of SSAS in inoperable patients.

With commercial approval of TAVR, the volume of these procedures in the United States has grown dramatically. Within Medicare, TAVR grew from 8.9% of all AVR procedures (across all risk groups) in 2012 to 50.3% in 2017.

 This rapid growth suggests a need for an analysis of the aggregate economic impact of TAVR treatment on the distribution of social welfare for inoperable patients; this type of study is referred to as social value analysis.

 These types of studies provide a useful complement to traditional cost-effectiveness analyses, as they incorporate the value to consumers, as well as the value to innovators, allowing policy makers to consider new technologies in a context that is both broad and deep. Inoperable patients who are treated with TAVR realize benefits in the form of increased quantity and quality of life relative to medical management. In addition, firms that manufacture TAVR devices realize benefits by way of profits earned on units sold. Payers may realize both costs (eg, via incremental new spending on TAVR devices) and benefits (eg, cost offsets realized through the avoidance of downstream events, such as postoperative inpatient readmissions). Assessing the value of TAVR to both payers and manufacturers allows for the estimation of the relative share of economic surplus returning to manufacturers (as opposed to consumers), therefore informing a broader discussion regarding optimal incentives for future innovation.

The purpose of this study was to estimate the present and future economic impact of TAVR adoption for inoperable patients with SSAS and for TAVR device manufacturers in the United States, using recently developed approaches for the assessment of economic value.

 Using an analysis time frame of 2018 to 2028, we aimed to project the size of the inoperable population with SSAS, estimate the fraction of this population who will receive TAVR, and calculate the economic returns of that utilization, for both the patients and the device manufacturers.

Projecting TAVR-Eligible Population From 2018 to 2028

To estimate the future economic impact of TAVR adoption for patients with inoperable SSAS and for TAVR manufacturers, we began by projecting the size of the inoperable population eligible for TAVR treatment.

To determine the number of inoperable patients with SSAS between 2018 and 2028, we first projected the number of patients with SSAS nationally, regardless of surgical risk status. We did this by multiplying current estimates of severe AS incidence rates (stratified by age) by the proportion of patients with severe AS whose disease is symptomatic,

 and then multiplying the resulting quantity by age-stratified US population projections for each year in the analysis period.

 Finally, we multiplied the values described earlier by the estimated fraction of patients with SSAS considered inoperable.

 The parameters used to perform these calculations are presented in 

Projecting TAVR Utilization From 2018 to 2028

Only a fraction of incident inoperable SSAS cases are treated with TAVR.

 As such, we estimated the number of inoperable patients using TAVR in the future as the product of (1) the projected eligible inoperable population in each year (described previously), and (2) an estimate of the fraction of eligible inoperable patients who ultimately receive TAVR, extracted from the literature.

 As a conservative assumption, we assumed that this fraction remains constant over time. We also assumed that individuals can receive the treatment only once.

TAVR market share estimates were available for the population with SSAS but not specifically for the inoperable population with SSAS. In addition, the prices of the different TAVR devices for the different SSAS risk groups (inoperable, high risk, and intermediate risk) are very similar.

 As such, for the purposes of this analysis, we used the best peer-reviewed cost-effectiveness evidence available, without regard to TAVR device type, and assumed that these results applied to all inoperable patients treated with TAVR.

 Cost-effectiveness results were then employed to calculate patient value (discussed later). As most contemporary TAVR procedures are performed using transfemoral (TF) access, we used cost-effectiveness results from a TF cohort.

We calculated industry revenue by multiplying the device price by the number of TAVR procedures on inoperable patients projected to occur annually between 2018 and 2028. We assumed that device cost would remain constant across years (Table 1

 In addition, we conservatively assumed profit—analogous to welfare gains in the context of social value analysis&mdash;to be 80% of revenue.

 This is substantially larger than the profit margins of 18% to 29% reported by device manufacturers in recent years.

 Formally, (undiscounted) industry value in year 

Computation of the Monetized QALY Benefit of TAVR 

We estimated the monetized value of the quality-adjusted life-year (QALY) gain for each treated inoperable patient by assuming a QALY willingness-to-pay (WTP) value of $150,000 (Table 1

 A threshold of $150,000 was chosen because incremental cost-effectiveness ratios (ICERs) above this cutoff are considered low economic value as defined by the American Heart Association and the American College of Cardiology.

 We estimated the total monetized QALY gain for the US inoperable SSAS population in each year as the product of the individual monetized QALY value and the number of TAVR procedures on inoperable patients in each year.

To circumvent the complexities of the US insurance market, this analysis used a societal perspective to estimate patient value. In this framework, we computed the net value that an individual obtains from TAVR as the difference between the monetized valuation of the QALY benefit (described previously) and the incremental difference in costs, including differences in the device cost, as well as differences in postprocedural medical expenses. The concept of patient value is thus an amalgamation of value returning to the patient (in the form of improved quality and/or quantity of life) and the change in medical costs borne by payers, which may be either patients themselves or, more typically, commercial or government insurers. For convenience, we refer to this construct as patient value. We extracted the values for cost offsets (including device cost) from the literature (Table 1

 and converted them to 2017 US$ using the Medical Consumer Price Index.

 We assumed a discount rate of 3% consistent with US guidelines.

 Formally, (undiscounted) patient value in year 

 denotes the QALY gain of TAVR relative to medical management, 

 denotes the estimated cost offsets of TAVR relative to medical management.

We assessed the model’s sensitivity to the joint uncertainty of all input parameters by implementing a probabilistic sensitivity analysis (PSA). In a PSA, model results are iteratively regenerated following resampling of input parameters from assumed statistical distributions, which are derived from published or assumed values (eg, means and CIs) describing the uncertainty of parameter point estimates. We varied QALY gains, lifetime cost offsets, incidence rates, the proportion of severe AS that is symptomatic, the proportion of SSAS that is inoperable, and the proportion of inoperable patients with SSAS who receive TAVR. Table 1

 shows the assumed statistical distributions of the uncertainty parameters and assumptions that governed the PSA. We performed 1000 replications of the model and computed the fraction of these in which estimated societal value exceeded estimated industry value. We also separately present results from a scenario analysis, which uses a WTP value of $100,000 per QALY in place of the base-case parameter value of $150,000.

Between 2018 and 2028, approximately 465,000 Americans with inoperable SSAS were projected to receive TAVR. Because the gradual aging of the US population will slowly increase the annual incidence of SSAS, regardless of risk status, the number of inoperable patients with SSAS receiving TAVR was also projected to grow over time. Full projection results are presented in 

In 2018, approximately 39,000 inoperable patients with SSAS were estimated to receive TAVR, resulting in a cumulative gain of about 50,000 QALYs (Table 2). After accounting for device costs and cost offsets, and using a WTP value of $150,000 per QALY, approximately $3.7 billion in patient value was created. By contrast, assuming that firms retain 80% of revenues as profits, approximately $940 million of the social value accrued to manufacturers. When these values were projected to 2028, we estimated that these undiscounted benefits would increase to $4.4 billion and $1.1 million ($3.2 billion and $810 million discounted) in patient value and manufacturer value, respectively.

Between 2018 and 2028, we projected the total QALY gain among inoperable patients with SSAS to be 600,486 QALYs and the cumulative social value of TAVR (summing over inoperable patients and manufacturers) to be $48.1 billion. Roughly 80% of that value ($38.5 billion) accrued to patients and roughly 20% ($9.6 billion) to manufacturers (Table 2). Furthermore, we projected that the undiscounted social value will grow over time as the incidence of SSAS increases (

Sensitivity and Scenario Analysis Results

 depicts repeated simulations of the model results following joint draws from the statistical distributions that describe the uncertainty in each of the model parameters (shown in Table 1

). The base-case (deterministic) result is indicated as a single light blue point; the results of each of the 1000 replications are represented as dark blue points. Points appearing above the orange line represent instances in which the value accruing to patients is greater than the value accruing to manufacturers, and points below it indicate the reverse. In 99.8% of the replicated simulations, the economic value to inoperable patients was greater than the economic value to manufacturers.

When the WTP for a QALY was assumed to be $100,000, the cumulative patient value was estimated to fall from $38.5 billion to $12.6 billion, and the estimated share of value accruing to manufacturers would rise from 20.0% to 43.3% (

). When the PSA was conducted using this lower valuation, patient value exceeded manufacturer value in 67.8% of the simulations.

Our findings suggest that substantial social value will be created as a result of TAVR treatment for inoperable patients with SSAS over the next decade. Our study projected that up to $48.1 billion in value will be generated between 2018 and 2028, with roughly 80% of that value accruing to inoperable patients and 20% to manufacturers. A PSA suggested that the portion of social value realized by inoperable patients is likely to exceed the value that will return to manufacturers, as this result was obtained in up to 99.8% of the simulated repetitions.

Prior research has found that TAVR is cost-effective compared with medical management in the inoperable population. Using data from the PARTNER B trial (ClinicalTrials.gov NCT00530894), Reynolds et al found that TAVR would generate a lifetime benefit of 1.29 QALYs per patient, at an incremental cost of $79,837, resulting in an ICER of $50,212 per QALY.

 Our study builds on these results by estimating the total social value that TAVR will generate and assessing the distribution of that value between patients and manufacturers. This perspective is useful for payers and policy makers because it extends the decision-making calculus around TAVR adoption beyond a narrow consideration of short-term costs. In addition, our study informs a larger debate in the literature regarding the relationship between the manufacturer share of economic returns from a new medical technology and incentives for future innovation.

 Although we observed a large social benefit of TAVR in the inoperable population with SSAS, it is possible that we underestimated the true benefit. First, the data used to inform our study were derived from the use of a first-generation device in the early age of TAVR. Over the past 10 years, there have been substantial improvements in device technology, procedural planning, operator experience, and post-TAVR care, which have been associated with better clinical outcomes and lower resource utilization.

 Another factor that may also contribute to underestimating the social benefit of TAVR in the inoperable population is the effect of prior undertreatment of this population on our future projection. Durko et al estimated that roughly one-third of inoperable patients with incident SSAS do not receive TAVR.

 This value aligns closely with our own comparison of the number of AVR procedures performed annually in the United States with our estimates of the number of annual incident cases (results not shown) and is consistent with previous research.

 Furthermore, previous research has suggested that roughly a quarter of patients with SSAS who are offered only SAVR will decline it.

 Accordingly, it is unknown what fraction of patients who are eligible for TAVR either are not offered it or decline it, thereby leading to the undertreatment of this population. To the extent that patient perception of TAVR as a less invasive alternative grows over time, undertreatment of SSAS may be reduced.

 As such, the cost-effectiveness of TAVR in the current era may be greater than previously estimated, and thus our calculations may underestimate the true social value of TAVR in the inoperable population with SSAS.

It is useful to consider our results in the context of other social value studies. Although this study provides the first assessment of the distributional effects on social value for TAVR, informative comparisons are available in the literature. Our estimate of the share of value returning to manufacturers for TAVR (roughly 20%) is nearly identical to the fraction estimated by a recent study on statins.

 A much lower estimate (2%) was obtained for pediatric vaccines,

 whereas another study found that between 25% and 66% of value accrues to manufacturers for 3 therapies treating multiple sclerosis, depending on modeling assumptions.

 This context suggests that, relative to several therapies, TAVR may generate reasonably large returns to inoperable patients and to society per dollar of manufacturer profit.

Of note, TAVR has also been demonstrated to be cost-effective relative to SAVR for operable patients at high

 surgical risk. Furthermore, recent clinical trials have demonstrated that TAVR yields substantial benefits relative to SAVR for patients at low surgical risk, who comprise the largest proportion of patients with SSAS.

 This suggests, first, that the social value of TAVR may extend beyond the inoperable risk group and, second, that this value will be inherently conditional on the degree to which patients and doctors select treatment with TAVR or SAVR. Future research should address these questions directly.

This study has several limitations. First, the results rely on projections of unknown future quantities (eg, the percentage of inoperable patients with SSAS who will receive TAVR). We sought to make realistic projections through analysis of data observed in the recent past. However, to the extent that our projections are inaccurate, our estimates may be biased. For example, we assumed the incidence of AS to remain constant, whereas real-world data show that the age-adjusted population incidence of AS decreased in recent years.

 Second, we estimated base-case results by incorporating several parameters obtained from the literature into a single economic model. Many of these parameters were estimated with uncertainty, including the QALY benefit and cost offset associated with TAVR usage, as well as the fraction of patients with SSAS considered inoperable. We sought to mitigate this limitation by incorporating the uncertainty associated with each input parameter into a joint PSA. Additionally, as noted previously, the estimated benefit of TAVR for inoperable patients (1.29 QALYs) was obtained from a cost-effectiveness study of an early-generation TAVR device (SAPIEN), as these are the only data available for this patient population.

 Because patient outcomes have improved over time and resource utilization has decreased, we may have underestimated the social value of TAVR by using this data. Third, the data that provided our QALY benefit and cost-offset input parameters considered only TAVR procedures conducted via the TF route

 and did not consider TAVR procedures performed via the less common alternative access routes (eg, transapical). Thus, our results cannot be generalized to TAVR performed via alternative access in this population, although we believe that inoperable patients eligible only for alternative access represent a small portion of the overall population, given that the rates of alternative-access TAVR have decreased substantially over time because of the progressively smaller delivery system of newer-generation TAVR devices.

 Finally, our estimate of the cost offsets associated with TAVR for inoperable patients comes from a study that used a 1-year period.

 Thus, our model did not account for any cost differences that might exist beyond this period.

We sought to estimate the value that will accrue to society—both in aggregate and to different segments of society&mdash;as a result of expected TAVR uptake by inoperable patients with SSAS between 2018 and 2028. We projected that up to $48.1 billion in social value will be generated, with roughly 80% of that value accruing to patients and the remainder accruing to manufacturers. Future research is needed to assess the social value of TAVR in patients with operable SSAS.

 Precision Health Economics, Oakland (JS, TTS) and Los Angeles (EvE, ABJ), CA; Lahey Hospital and Medical Center (SJB), Burlington, MA; Edwards Lifesciences (CT, SC), Irvine, CA; Harvard Medical School (ABJ), Boston, MA; Massachusetts General Hospital (ABJ), Boston, MA.

 Dr Sussell was an employee of Precision Health Economics, a consultancy of the health and life sciences industries, at the time of this study. Ms van Eijndhoven and Mr Schwartz are employees of Precision Health Economics. Dr Baron reports consulting for Edwards Lifesciences, advisory board membership for Abiomed and Boston Scientific Corp, and a pending research grant from Boston Scientific Corp. Dr Thompson and Mr Clancy are full-time employees of and own stock in Edwards Lifesciences, which manufactures transcatheter aortic valve implants. Dr Jena has received consulting fees unrelated to this work from Pfizer, Bristol-Myers Squibb, Novartis, Amgen, Eli Lilly, Vertex Pharmaceuticals, AstraZeneca, Celgene, Tesaro, Sanofi Aventis, Biogen, and Precision Health Economics and has received consulting fees for economic expert testimony from Analysis Group.

 Concept and design (JS, EvE, TTS, SJB, CT, SC, ABJ); acquisition of data (JS, EvE, TTS, SC); analysis and interpretation of data (JS, EvE, TTS, SJB, CT, SC, ABJ); drafting of the manuscript (JS, EvE, TTS, SJB, SC, ABJ); critical revision of the manuscript for important intellectual content (JS, EvE, TTS, SJB, CT, SC, ABJ); statistical analysis (JS, EvE, ABJ); provision of patients or study materials (CT, SC); obtaining funding (SC); administrative, technical, or logistic support (TTS, CT, SC); and supervision (JS, CT, SC, ABJ).

 Emma van Eijndhoven, MS, MA, Precision Health Economics, 11100 Santa Monica Blvd, Ste 500, Los Angeles, CA 90025. Email: emma.vaneijndhoven@precisionhealtheconomics.com.

Articles

Transcatheter aortic valve replacement for inoperable severe, symptomatic aortic stenosis will create significant social value in the next decade, mostly accruing to patients versus manufacturers.

Economic Value of Transcatheter Valve Replacement for Inoperable Aortic Stenosis

 The causes of oncology drug price growth remain unclear. Analyzing corresponding trends in revenue can help understand these causes. This study seeks to assess changes over time in prices, patient counts, and drug-level revenues in the US market for oncology therapies and to investigate whether price growth is driven by an increased ability by pharmaceutical firms to capture profits.

 Nineteen-year retrospective study (1997-2015).

 We used panel regression to investigate trends in prices, patient counts, and revenues within a US national data set consisting of targeted oncology therapies launched in different eras.

 We find that prices have roughly tripled, whereas average patient counts per therapy have fallen by 85% to 90% over this period. However, the entire distribution of annual revenues has fallen: For instance, median revenues for drugs launched in the early 2010s are about half of what they were for drugs launched in the late 1990s.

 Future research on the causes of quantity decline can help inform pharmaceutical policy.

Comparing targeted oncology therapies launched in the intervals 1997-2002, 2003-2009, and 2010-2015, we investigate trends over time in prices, the number of patients using each drug, and revenues within the US market.

We confirm the previously documented result that prices are rising over time.

However, revenues are not rising concurrently with prices.

This decline is attributable to contemporaneous declines of 85% to 90% in per-drug average patient counts.

Insurers, healthcare providers, and others have noted that prices for novel oncology treatments are rising over time.

 Price per episode of treatment has risen steadily, and prior research suggests that although the efficacy of cancer therapies (as measured by gains in overall survival) is certainly improving over time, prices are rising even faster.

 If improved efficacy or effectiveness does not provide the dominant explanation, then what else might be contributing to the observed increase in prices?

One possibility is that firms today possess greater market power and use it to earn greater rewards for their drugs. If true, higher prices would coincide with higher revenues. At first blush, the magnitude of price growth in the oncology market may make growth in revenues seem self-evident. However, it becomes more difficult to assess when one considers the fact that newer drugs tend to treat fewer patients, as we document later. This decline in patients treated may be due to greater personalization of therapy, slower incremental progress that fails to force older drugs out of the market, or other factors. Regardless, rising prices alongside falling quantities make the trend in revenues an empirical question.

An empirically interesting context for this investigation is the market for targeted oncology therapies, first introduced in the late 1990s. Prior research has documented that targeted therapies have quickly come to dominate the market: For example, targeted therapies accounted for about two-thirds of all chemotherapy expenditures by 2011.

 In this study, we investigate trends in prices, quantities, and revenues for targeted oncology therapies. We aim to determine whether higher prices have coincided with higher revenues and rewards for innovation or whether, instead, price growth has coincided with flat or falling revenues. Analyzing a sample of targeted therapies intended to treat common tumor types, we estimate the growth in price per patient-year, the reduction in the average size of the annual patient base using each drug, and the resulting change in revenue. In auxiliary analysis, we also estimate the implied increase in costs of drug development per patient-year.

We focused on targeted cancer drugs launched since 1997, when the first targeted agent (monoclonal antibody) was approved by the FDA. We considered all drugs primarily aimed at extending survival and/or progression-free survival for patients with cancer and focused on the 6 most common tumor types: breast, colorectal, melanoma, non-Hodgkin lymphoma, non—small cell lung, and prostate.

 As some drugs are indicated for more than 1 tumor type, the unit of analysis for this study was therapy—tumor pair. To account for the fact that many drugs have more than 1 indication, and that approvals for new indications are frequently granted subsequent to the initial approval, we computed estimates for all possible therapy–tumor pairs, including all indications known at the time of this analysis. We estimated relative usage of a single agent for different indications by assuming usage proportional to relative disease incidence. After we dropped data for indications not of interest, our final sample consisted of 29 therapies and 33 therapy–tumor pairs (see 

Using data for each of these therapy—tumor pairs, we defined 3 measures of interest: therapy price, the number of patients using each drug for each indication, and annual revenues. We estimated each of these quantities for each therapy–tumor pair of interest, for each year in our analysis sample (1997-2015), and then used panel regression to investigate how they changed with respect to drug launch year. In supplementary analysis presented in the eAppendix, we also estimate trends in per-patient research and development (R&D) cost.

Therapy Price, Patient Count, and Revenues

We estimated the total number of patients and the estimated price of treatment using the IQVIA National Sales Perspective (NSP) data set.

 The NSP reports nationally representative estimates of the total annual units of individual therapies distributed in the United States and of the total related revenues received by manufacturing firms. We combined these data with monthly dose and average length of treatment values specific to each therapy—tumor pair (obtained from FDA labels and clinical trials) to derive our price and patient count outcomes. Price was calculated as the cost for a full course of treatment (ie, the total revenue estimated to return to the manufacturer as a result of 1 patient being treated). Patient counts were imputed through analysis of the total number of therapy units distributed in combination with label information on dosing for a full course of treatment. Revenues were calculated within-year as the product of price and patient count. Revenues were calculated at the therapy, rather than therapy–tumor, level because this captures the total return on launching a new drug. Complete details on the construction of these measures are provided in the eAppendix.

To validate our findings regarding trends over time in the average number of patients, we separately analyzed patient counts in an independent data set, the Medicare Current Beneficiary Survey (MCBS). The MCBS is a survey of a representative sample of Medicare beneficiaries and contains data on diagnosis and drug utilization.

 For each of the therapy—tumor pairs of interest, we counted the number of individuals in the MCBS sample who (1) had an indication for that tumor type and (2) reported using that therapy. We then used sampling weights to inflate these counts to the population level. We note that the MCBS results are estimates of counts for the Medicare population only and should not be construed as separate estimates for the aggregate US population. We repeated this process for the individual years between 1997 and 2012. (At the time of this analysis, 2012 was the most recent year for which MCBS data were available.) Finally, we compared trends over time in average patient counts as estimated in the IQVIA and MCBS analyses.

We sought to determine whether each of our main outcomes (therapy price, number of patients, and annual revenues) was correlated with therapy launch year. To do this, we fit a series of regression models using the above measures as outcomes and therapy launch period as the key independent variable. We converted therapy launch year into a categorical variable with 3 groups: drugs launched between 1997 and 2002 (reference category), drugs launched between 2003 and 2009, and drugs launched between 2010 and 2015. We present results from 2 sets of regressions. In the first, we regress each of our outcomes of interest on this categorical variable corresponding to launch time period. In the second, we include an additional regressor (years post launch) to control for life cycle trends in the price and quantity of a drug following its market entry. We report trends in average regression-adjusted price and quantity and then report movement over time in the entire distribution of regression-adjusted annual revenues per therapy. We conducted sensitivity analyses involving (1) adding covariates to the base-case model, (2) using alternative values for key parameters related to the cost of R&D, and (3) asserting a linear relationship between launch year and outcomes, rather than the period fixed-effects structure described above. These are discussed in detail in the eAppendix.

We begin with an analysis of the first study outcome, episode treatment price. The fact that newer oncology products are increasingly costly has been extensively documented in the literature.

 This trend is also readily apparent in our data on targeted agents. 

 presents the results of a series of regressions that use therapy price as the dependent variable and launch year as the key independent variable and that follow the specifications defined previously.

We find a statistically significant positive correlation between price and launch year. The episode treatment cost for drugs launched between 2003 and 2009 was, on average, $23,000 greater than that for drugs launched between 1997 and 2002. The difference was even greater for drugs launched between 2010 and 2015—a statistically significant average difference in episode treatment cost of about $43,000.

This study sought to examine whether price growth has coincided with revenue growth. Thus, trends in annual average patient counts play a crucial role.

 presents the results of the regressions for the patient count outcome, first without and then with the control for time since launch. Figure 2 demonstrates a strong negative relationship between launch year and average patient count. Relative to therapies launched in the early period, the dummy models suggest that therapies launched in the middle period were used by, on average, 28,000 to 35,000 fewer patients annually, whereas therapies launched in the late period were used by 33,000 to 44,000 fewer patients. Detailed time-series plots of patient counts for individual therapy—tumor pairs, and for average values within launch period, are presented in 

To confirm these results, we conducted a separate analysis of patient counts by therapy—tumor pair in the independent MCBS data set. Comparative results are presented in 

Note that the IQVIA data set covers the entire US population, whereas MCBS covers Medicare patients only. Each is designed to be nationally representative for its particular sample frame.

There is clear evidence that annual patient populations are smaller for more recently launched drugs: In the main analysis using IQVIA data, the average patient count fell from 48,520 per drug for drugs launched in the early period to 4781 per drug for drugs launched in the late period, a decline of 90%. A decline of similar magnitude (85%) is observed in the Medicare data.

The reduction in quantity seems to have offset growth in price. The entire distribution of annual revenues has fallen over time. We use the regression-adjusted (ie, predicted) revenues from our regression model of revenues as a function of years since launch and time period. We also aggregate this up to the therapy level to eliminate the possibility that newer drugs spawn more indications and thus artificially lower revenues per tumor type. This permits uniform comparisons over time that account for the way in which revenue evolves over the life cycle of a drug. 

 (A and B) presents the distribution of regression-adjusted annual revenues (at the therapy level) for each of the 3 launch periods; the difference between the 2 panels is that Figure 4B removes a single influential outlier—bevacizumab&mdash;from the data set. Both panels show that the distribution of regression-adjusted annual revenue has shifted left over time. In both cases, the most recent distribution ranges from $250 million to $500 million, whereas the earliest period shows a distribution from about $250 million to more than $900 million (all values are reported in 2015 US$). The sole difference between the distributions lies in the middle period. In the full sample, the main mass of the distribution lies between those of the early and late launch periods, but significant right skew is present. This long right tail is caused by the presence of a single drug, the blockbuster bevacizumab. Annual revenues for that drug (limited to the 6 indications of interest to this study) routinely exceeded $1 billion, in part because it was approved for more than 1 of those tumor types.

 presents the shifts in the distributions at key percentile points. Because bevacizumab is such an outlier, the Table accurately describes the distributions in both panels of Figure 4 (ie, the bevacizumab data points lie beyond the 90th percentile).

Annual adjusted revenues for the median drug have fallen from about $580 million to $287 million, a decline of about 50%. There is a decline of roughly 40% at the 25th percentile and nearly 60% at the 75th percentile. The only region of increase occurs at the 10th percentile, where revenues increased from the early to middle period, only to fall back down in the final period.

Results from the sensitivity analyses are presented and explained in the eAppendix.

The rising cost of novel oncology therapies has been a source of great controversy in recent years.

 Our analysis confirms that prices of oncology drugs are indeed rising rapidly. We show that the number of patients taking each drug has dropped substantially over the same period of time. As a result, revenues have fallen at every point in the distribution, after accounting for life cycle growth in revenues over years since launch. This suggests that price growth is unlikely to have resulted from greater pricing power, at least within this market segment. Profit-maximizing firms with more pricing power would never willingly make decisions that lead to lower revenues for each drug launched. One exception to this point might occur if costs of drug discovery or production have fallen significantly. Although the extant academic literature on the costs of drug discovery remains controversial, all of it points to rising costs.

 We know of no academic publications on trends in the costs of oncology drug production; more research is called for in this area.

This study has several important limitations. The IQVIA NSP data set, which provided much of the source data for this study, contained “restrictions” (partially missing values) for some therapies in some years; we recoded these as missing. This step will not affect the results of the study, as long as the circumstance of missing data is not correlated with our outcomes of interest. Also, our measure of the cost of individual therapies contains only the component of total cost that returns to pharmaceutical manufacturers; it does not incorporate markups by wholesalers or hospitals, nor does it include any confidential rebates paid by manufacturers to purchasers. In addition, we focus on only treatments for the 6 most common tumor types; as such, our results have limited generalizability to other forms of cancer. Finally, to estimate therapy prices, we assume that average treatment duration is equal to the duration indicated on the drug label. In practice, individual patients’ duration of therapy may be longer or shorter than is suggested by the label because of factors such as mortality, discontinuation, and extended treatment at the discretion of the physician.

Previous research has suggested that we should be skeptical of the notion that prices are rising solely because effectiveness is rising.

 Our study also casts doubt on an explanation for price growth that relies solely on rising R&D costs: If this were so, firms would necessarily respond by launching drugs capable of earning higher revenues. Our findings suggest instead a relationship between price growth and average patient counts, although the precise nature of this relationship is not fully clear. One possible explanation for declining patient counts is relatively slow growth in effectiveness over time; this would improve the ability of older drugs to remain on the market. The consequences of this longevity would be reduced market share and reduced revenue for newer drugs. A second possible hypothesis for declining patient counts would be increased competition: All else being equal, an increase in the number of drugs approved for a given tumor type would lead to a decline in the average number of patients per drug. This explanation, however, would also suggest an increase in price competition, which is inconsistent with the observed data.

A final possible explanation for the trends we observe is growth in the development of drugs that target patients with specific biomarkers (sometimes referred to as personalized medicine) within the targeted oncology market. By design, these drugs target subsets of the total population of patients with the indicated cancer. Shrinking patient counts might be in part the result of more personalized therapies that treat narrower indications. For example, trastuzumab is indicated for human epidermal growth factor receptor-2—overexpressing breast cancer, which comprises only 15% to 20% of invasive breast cancer cases.

 Mechanically, personalization leads to lower revenue when we hold prices constant. From another perspective, this is a cost to society of personalization: Without significant offsetting price growth, personalized therapies generate lower revenues and returns to innovators; this reduction in returns may reduce the rate of drug discovery in the long run.

 More research is needed on these and other hypotheses for the causes of drug price growth and the observed decline in average patient counts.

Despite acknowledged limitations, this study provides surprising new data on declining patient populations treated by targeted cancer agents. This pattern is likely an important and, to our knowledge, previously undescribed factor that lies behind trends in revenues and rewards from innovation in oncology.

Precision Health Economics (JS, JV-O, DV, DL), Oakland, CA; Genentech (HPG), South San Francisco, CA.

 Dr Vanderpuye-Orgle is an employee of and owns stock in Amgen, Inc, which manufactures oncology therapies. Mr Goertz is employed by Genentech/Roche and owns Roche stock; Genentech manufactures and sells several of the oncology therapies covered in this manuscript. Dr Lakdawalla is a consultant to Precision Health Economics, a healthcare consultancy with clients in the life science and insurance industries, including firms that market oncology treatments; he holds equity in Precision Medicine Group, parent company of Precision Health Economics. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (JS, JV-O, HPG, DL); acquisition of data (JS, DV); analysis and interpretation of data (JS, JV-O, HPG, DL); drafting of the manuscript (JS, DV); critical revision of the manuscript for important intellectual content (JS, JV-O, HPG, DL); statistical analysis (JS); provision of patients or study materials (HPG); obtaining funding (HPG); administrative, technical, or logistic support (JS, DV); and supervision (JS, HPG, DL).

 Jesse Sussell, PhD, Precision Health Economics, 555 12th St, Ste 1725, Oakland, CA 94607. Email: jesse.sussell@pheconomics.com.

1. Chhatwal J, Mathisen M, Kantarjian H. Are high drug prices for hematologic malignancies justified? a critical analysis. 

. 2015;121(19):3372-3379. doi: 10.1002/cncr.29512.

2. Lotvin AM, Shrank WH, Singh SC, Falit BP, Brennan TA. Specialty medications: traditional and novel tools can address rising spending on these costly drugs. 

. 2014;33(10):1736-1744. doi: 10.1377/hlthaff.2014.0511.

3. Kantarjian H, Rajkumar SV. Why are cancer drugs so expensive in the United States, and what are the solutions? 

. 2015;90(4):500-504. doi: 10.1016/j.mayocp.2015.01.014.

4. Howard DH, Bach PB, Berndt ER, Conti RM. Pricing in the market for anticancer drugs. 

The prices of targeted oncology therapies have grown substantially, but revenues have not. This is due in part to large declines in per-drug patient counts.

Understanding Price Growth in the Market for Targeted Oncology Therapies

 To illustrate a more comprehensive view of value associated with medicines treating a highly severe illness and to apply these insights to estimate the costs and benefits of 3 treatments for multiple sclerosis (MS): Avonex, Tysabri, and Tecfidera.

 Retrospective study spanning 2002 to 2013. We used economic theory to derive the value of therapy to patients with MS and to individuals who face the risk of contracting MS in the future, under the alternative assumptions that therapies were fully insured or paid for out of pocket.

 Models were parameterized through secondary data analysis and targeted literature review. Estimates of individual value were aggregated to the societal level using therapy-specific treatment prevalence rates. Aggregate consumer value was compared with manufacturer revenue.

 In the baseline model, Avonex, Tysabri, and Tecfidera generated $46.2 billion of total value to consumers, almost one-third of which accrued to those without MS. The total value to consumers was double manufacturer revenue. Results were qualitatively robust to the use of alternate epidemiological and economic parameters. We found that value to the healthy is positively related to disease severity, and that value to both the sick and the healthy are larger when costs are shared via health insurance.

 Theory predicts that treatments for severe disease provide “peace of mind” value to the healthy. Avonex, Tysabri, and Tecfidera have generated significant social value, a large majority of which accrues to consumers. Future economic valuations of medical technology should consider both the potential value to the healthy and the effects of insurance.

	Although many studies have assessed the social value of medical care to the sick, the value to the healthy who may use treatment if they become sick has been largely ignored. We used empirical estimations to parameterize an economic model that describes the value of 3 multiple sclerosis treatments to those who are healthy but face the risk of contracting MS in the future, as well as to the sick. 

		When patients bear the full cost of treatment, the value of the 3 treatments to the sick totals $11.1 billion, while the value to the healthy is $8.9 billion. 

		The value of therapy increases with the severity of the disease being treated. 

		Insurance coverage has a complementary effect on the value of therapy: the total populationwide value of the 3 treatments increases to $46.2 billion when actuarially fair insurance is assumed.

The rising cost of healthcare in recent decades has been accompanied by an increasing interest in quantifying the value of medicine.

 The cost of healthcare—unlike the costs of other goods&mdash;is often borne primarily by healthy consumers who are not currently using it. For example, premiums for private health insurance and taxes for public health insurance schemes are paid by the entire population, not just the patients who happen to be sick at a point in time. This raises a key question: What is the value of healthcare to the sick and to the healthy consumers who are paying for it?

The value of medical care to the sick is readily apparent, whereas the value to the healthy manifests in at least 2 ways. First, healthy individuals value medical technology because it will be available to them if they become sick in the future. The higher the likelihood of becoming sick with a particular disease, the more a healthy person values treatments for that disease. Second, new technology can provide peace of mind in the present, even to those who may never end up using it. The worse the disease, the more a healthy person values this peace of mind.

To illustrate the “peace of mind” value, consider the analogy of a fire extinguisher in one’s home: it provides value should the house catch fire by reducing the damage the fire would cause, and this value increases with the size of the potential loss. Awareness of this potential benefit provides immediate and continuous peace of mind from the protection against fire damage. This value is realized even if a fire never breaks out. Moreover, this peace of mind value increases with the size of the potential loss. A renter with few possessions may worry little about the risk of fire and thus derive few benefits from a fire extinguisher. In contrast, the owner of a home filled with priceless heirlooms might worry more and thus place higher value on the fire extinguisher. The “peace of mind” value is likely to be quantitatively meaningful, because evidence suggests that most consumers dislike risk and value reducing it.

Medical technology provides peace of mind similar to that in the fire extinguisher example. To illustrate, consider a healthy individual today and another in 1990 who are concerned about the prospect of a human immunodeficiency virus (HIV) infection. Both would experience anxiety, but the first individual would be anxious about the risk of complications and the inconvenience of a lifetime of medical treatment; the second individual would be anxious about death. The very real difference between these 2 levels of anxiety contributes to the value that healthy individuals have obtained from modern HIV and AIDS therapies. In other disease areas, the value of a new therapy to a healthy individual can be similarly characterized by decreased anxiety or fear of a diagnosis due to the therapy’s ability to reduce the harm from a disease. This example illustrates that the peace of mind afforded by new medical technologies will be especially valuable for treatments that mitigate the consequences of the most severe diseases.

Multiple sclerosis (MS) provides a useful case study of a severe disease, as it is the leading cause of nontraumatic neurologic disability among young adults.

 In MS, the body’s immune system attacks the central nervous system, creating brain lesions. During relapses, symptoms dramatically worsen and the disease can transition into a stage of progressive disability. MS patients suffer from fatigue and pain, as well as mobility and sensory problems.

 Peak onset occurs between the ages of 20 and 40, often affecting healthy individuals in their prime years of productivity. Thus, MS onset imposes high medical costs and has severe consequences for quality of life and productivity, such as lost income.

MS therapies also highlight the wider debate over the value of new medical technology. Some question the value of innovative drugs that help manage—but do not cure&mdash;debilitating and progressive diseases. Skepticism about the value of such drugs has been fueled by cost-effectiveness studies and a recent United Kingdom risk-sharing scheme.

Our study uses an economic model developed by Lakdawalla, Malani, and Reif (2015)

 to estimate the value of MS therapies to both healthy and sick individuals. We focus on 3 currently available MS therapies, incorporating their specific dates of introduction, magnitudes of health benefit, and prices: Avonex (interferon beta-1a intramuscular, introduced 1996), Tysabri (natalizumab, introduced 2004), and Tecfidera (dimethyl fumarate, introduced 2013).

From an economic perspective, the value of a good is measured as the amount of other consumption that an individual is willing to sacrifice in exchange for it. These trade-offs are conventionally estimated in the framework of a “utility” model that explicitly estimates the value that consumers assign to different goods. A plethora of studies measure the value consumers assign to health relative to other goods,

 and these measurements provide the empirical basis for utility models that estimate the trade-off between health improvements and other consumption. We followed this approach and estimated the value of the MS therapies of interest by constructing an economic model of the trade-off between consumption and health. Following the economic literature, we assumed that the quantity of consumption is determined by the income that remains after medical costs.

 As we describe above, value may accrue not only to those suffering from an illness, but also to those who are currently healthy but still susceptible to future illness. We refer to these constructs as “value to the sick” and “value to the healthy.” Both of these depend, in turn, on how the costs of therapy are incurred. Unlike standard goods, a portion of healthcare is often paid for by nonusers, via insurance. Insurance may increase the value of therapy for both the sick (by replacing direct costs with less costly insurance premiums) and the healthy (by reducing financial risk). Our study thus estimates the value of MS therapy from 4 perspectives: value to the sick and value to the healthy, first under the assumption that costs are fully borne by consumers (without insurance) and then assuming actuarially fair insurance, in which therapy costs are distributed across the entire risk pool (with insurance).

Prior efforts to estimate the value of new medical technologies have typically emphasized only 1 of these 4 perspectives: the value to the sick, without consideration of insurance.

) formally describes the economic model developed by Lakdawalla, Malani, and Reif (2015),

 which we used to measure value from each of these perspectives (utility model—based willingness-to-pay estimation). The value to the sick depends on: 1) health benefits of the therapy for those who are diagnosed with MS, as measured by incremental quality-adjusted life-years (QALYs); 2) the health costs of MS, as measured by QALYs for individuals with and without MS; 3) the costs of therapy; 4) other medical costs with and without therapy; and 5) differences in consumer income, which determine the value of money to a consumer. We measured these both for MS patients receiving best supportive care (BSC) and for MS patients utilizing 1 of the 3 qualified drugs.

The value to the healthy depends on all 5 factors above, along with: 6) the incidence of MS, which measures the risk that healthy people will acquire the disease in any given year; and 7) the degree of consumer risk-aversion, which measures the value of risk-reduction to healthy consumers. In reality, some individuals are not materially at risk for MS, meaning that the population could consist of 3 groups: those with MS, healthy individuals at risk for MS, and healthy individuals not at risk for MS. This third group may never derive benefit from the actual use of the therapies. However, as the causes of MS are still not well understood,

 healthy individuals cannot easily ascertain whether they fall into the second or third groups. Thus, for the purposes of our analysis, we pooled these groups together.

We used parameters in these 7 areas to construct separate economic utility models for each of the 4 perspectives described in Figure 1. Our models assume that the health state and drug utilization choice are constant for an individual within each year. These models are then used to estimate the annual value to consumers of using 1 of our 3 drugs of interest relative to BSC. The incremental value of the 3 drugs is given by the difference in value between using the drug and using BSC.

Finally, we aggregated the different estimates of incremental per-patient value to the societal level using disease prevalence and drug utilization rates. We added up the individual annual values of treatment over the period 2002 to 2013 (the years for which data on the therapies are available) to obtain the aggregate value of the 3 therapies. These aggregate values have been compared with manufacturer revenue to determine the share of value that returns to consumers. Complete details on economic model specification, parameterization, and sensitivity analyses are provided in the eAppendix.

 summarizes the parameters obtained from our literature review and data analysis, which were used to calculate the social value of therapies. As detailed in the eAppendix, our analysis suggests that MS patients earn 37.1% less income than their non-MS counterparts; Avonex users earn 41.4% more income than their MS patient counterparts who are not using disease-modifying therapies (DMTs). Because of data limitations, we were unable to estimate income effects for Tysabri and Tecfidera directly. Instead, we assumed that the effects for those therapies were equal to that of Avonex. This conservative assumption likely understates the income effect of those therapies, because both of those products reduce relapse rates and disability progression more than Avonex does.

An MS diagnosis was also associated with a significant increase in non-DMT medical costs (87.0%), while the use of Avonex and Tysabri reduced annual medical costs by 11.2% and 16.4%, respectively. There were too few cases of Tecfidera usage in the claims data to identify an effect on medical costs (Tecfidera had a sample size of 137 compared with 9272 and 1223 for Avonex and Tysabri, respectively). As a result, we elected to use the Avonex cost offset parameter (—11.2%) for Tecfidera. This is a conservative approach, as Tecfidera was shown to reduce disability progression and relapse frequency more compared with interferons (including Avonex).

As a result of using this conservative estimate, our models likely underestimate the social value of Tecfidera. In the eAppendix, we describe a sensitivity analysis in which this value is set equal to the midpoint of the Avonex and Tysabri estimates (—13.8%).

To complete the economic model, we used established estimates from the literature for the health and risk-aversion parameters. The Table summarizes the baseline and sensitivity values used for the MS epidemiological parameters,

 and the economic parameters for the value of health.

 In addition, the Table displays values used in sensitivity analyses that account for insurance loading and, separately, the cost of drug development borne by manufacturers.

 provides baseline estimates of value for all 3 drugs combined, aggregated over all years from 2002 through 2013. Aggregate value to the sick, when bearing the full cost of therapy, is estimated to be $11.1 billion. When actuarially fair insurance is available—so that the healthy and sick share the cost of treatment&mdash;value to the sick almost triples, to $31.8 billion. Conversely, value to the healthy without insurance is estimated to be $8.9 billion. When insurance is available, value to the healthy rises to $14.4 billion. This increase demonstrates the value of financial risk reduction that is obtained with insurance coverage.

Based on an 80% national average insurance coverage rate, the total value of the 3 therapies is estimated to be $40.9 billion.

 Overall, these results suggest that estimates of the value of medical technologies which ignore either the benefits that accrue to the healthy or the role of health insurance may be biased downward—perhaps severely so.

Impact of Disease Severity on Value to an Individual Insurance Enrollee

Conceptually, the value to the healthy should be higher when considering treatments for more severe diseases. For instance, an effective new treatment for a highly fatal disease provides significantly more peace of mind to the healthy than one for a mild skin condition. Our analysis confirms this intuition by re-estimating the value of 1 therapy (Tysabri) to a healthy individual with insurance, while incrementally varying the assumed severity of MS, holding other factors (including the absolute treatment effect) constant. The results are depicted in 

If MS was not a severe disease, the value of Tysabri to a healthy individual would be small. This is evident on the left side of Figure 3: as the assumed QALY of untreated MS approaches 1, the value of treatment to a healthy individual approaches 0. However, MS is a debilitating disease, with an estimated untreated QALY value of 0.584 for those patients who might be treated by Tysabri.

 At this level, our model estimates the monthly value of Tysabri to a healthy individual to be $6.26. By contrast, we calculate that the actuarially fair per-member-per-month cost of insurance coverage of Tysabri is an order of magnitude smaller—about $0.48. This suggests that individual insurance enrollees gain more value from access to coverage than they lose due to the associated incremental insurance premium.

Significantly, the value of the treatment varies with disease severity, even when clinical effectiveness is held constant. Intuitively, a given improvement in clinical status is worth more to a patient suffering from a more severe disease. Therefore, singular focus on efficacy and/or effectiveness may ignore an important additional determinant of value.

 portrays the relative share of lifetime value accruing to all consumers (both healthy and sick) and manufacturers, aggregated across the 3 therapies. When no insurance is available, an estimated 49% of value accrues to consumers ($20.0 billion—the “population-wide value” previously described), and 51% accrues to manufacturers as revenues ($21.2 billion). Because most individuals in the United States had health insurance during the time period of the study,

 the values under full insurance are empirically relevant. When full insurance is assumed, the share of value accruing to consumers rises to 69% ($46.2 billion), with the remaining 31% accruing to manufacturers. We conservatively assumed that all revenues ($21.2 billion) accrue to the manufacturer as profits. In reality, costs are not 0, and as a result, the true share of value accruing to consumers will be larger than what we have estimated here. In the sensitivity analyses, we provide a revised estimate of the distribution of surplus that incorporates estimates of the opportunity cost of research and development.

Our model relies on both epidemiological (eg, incidence rate) and economic (eg, risk aversion) inputs, obtained from the literature and from our original data analysis. Varying these inputs moderately alters the results presented above. For example, assuming the availability of health insurance, estimates of the share of overall value accruing to consumers range from 59% (when the relative value of health is reduced) to 75% (when the treatment prevalence rate is reduced). When using the lower bound for risk aversion (0.15), the share to consumers (assuming insurance coverage) is 62%. These results are presented in the eAppendix (exhibits A12 and A13).

In addition to relying on parameters retrieved from the literature, our model takes as inputs parameters obtained via novel data analysis—specifically the income and medical cost effects of MS (relative to no MS) and of therapy (conditional on MS). These parameters have associated error distributions, and we accounted for these distributions using bootstrap methods. We created 1000 weighted bootstrap samples from the Medical Expenditure Panel Survey and claims datasets and estimated the parameter of interest (population-wide value&mdash;the sum of aggregate value to the sick and aggregate value to the healthy) from each set of regression results. The results are qualitatively robust to the introduction of these error distributions: 95% of the resampled estimates show more aggregate value accruing to consumers than to the manufacturer. The distribution of bootstrapped estimates is presented in the eAppendix (exhibit A14).

At baseline, our model assumes that actuarially fair insurance is available; however, in reality, insurance always involves some loading cost to cover administrative overheads.

 We therefore conducted a sensitivity analysis using an administrative load parameter of 16% (the median of the values reported by Karaca-Mandic et al [2011]).

Availability of multiple sclerosis (MS) therapies provides substantial value to the currently healthy (who may contract MS in the future), particularly when treatment is fully covered by insurance. 

Jesse Sussell, PhD

Articles

Economic Value of Transcatheter Valve Replacement for Inoperable Aortic Stenosis

Understanding Price Growth in the Market for Targeted Oncology Therapies

Reconsidering the Economic Value of Multiple Sclerosis Therapies