Key PointsQuestion
How does National Institutes of Health (NIH) investment in pharmaceutical innovation compare with investment by the pharmaceutical industry?
Findings
In this cross-sectional study of 356 drugs approved by the US Food and Drug Administration from 2010 to 2019, the NIH spent $1.44 billion per approval on basic or applied research for products with novel targets or $599 million per approval considering applications of basic research to multiple products. Spending from the NIH was not less than industry spending, with full costs of these investments calculated with comparable accounting.
Meaning
The results of this cross-sectional study suggest that the relative scale of NIH and industry investment in new drugs may provide a basis for calibrating the balance of social and private returns from these products.
Importance
Government and the pharmaceutical industry make substantive contributions to pharmaceutical innovation. This study compared the investments by the National Institutes of Health (NIH) and industry and estimated the cost basis for assessing the balance of social and private returns.
Objectives
To compare NIH and industry investments in recent drug approvals.
Design, Setting, and Participants
This cross-sectional study of NIH funding associated with drugs approved by the FDA from 2010 to 2019 was conducted from May 2020 to July 2022 and accounted for basic and applied research, failed clinical candidates, and discount rates for government spending compared with analogous estimates of industry investment.
Main Outcomes and Measures
Costs from the NIH for research associated with drug approvals.
Results
Funding from the NIH was contributed to 354 of 356 drugs (99.4%) approved from 2010 to 2019 totaling $187 billion, with a mean (SD) $1344.6 ($1433.1) million per target for basic research on drug targets and $51.8 ($96.8) million per drug for applied research on products. Including costs for failed clinical candidates, mean (SD) NIH costs were $1441.5 ($1372.0) million per approval or $1730.3 ($1657.6) million per approval, estimated with a 3% discount rate. The mean (SD) NIH spending was $2956.0 ($3106.3) million per approval with a 10.5% cost of capital, which estimates the cost savings to industry from NIH spending. Spending and approval by NIH for 81 first-to-target drugs was greater than reported industry spending on 63 drugs approved from 2010 to 2019 (difference, −$1998.4 million; 95% CI, −$3302.1 million to −$694.6 million; P = .003). Spending from the NIH was not less than industry spending considering clinical failures, a 3% discount rate for NIH spending, and a 10.5% cost of capital for the industry (difference, −$1435.3 million; 95% CI, −$3114.6 million to $244.0 million; P = .09) or when industry spending included prehuman research (difference, −$1394.8 million; 95% CI, −$3774.8 million to $985.2 million; P = .25). Accounting for spillovers of NIH-funded basic research on drug targets to multiple products, NIH costs were $711.3 million with a 3% discount rate, which was less than the range of reported industry costs with 10.5% cost of capital.
Conclusions and Relevance
The results of this cross-sectional study found that NIH investment in drugs approved from 2010 to 2019 was not less than investment by the pharmaceutical industry, with comparable accounting for basic and applied research, failed clinical trials, and cost of capital or discount rates. The relative scale of NIH and industry investment may provide a cost basis for calibrating the balance of social and private returns from investments in pharmaceutical innovation.
Private sector investment and returns are classically viewed as the primary driving force for innovation. Evidence also shows that public sector investments in basic and applied biomedical research, including those from the National Institutes of Health (NIH), contribute substantively to the emergence of new drugs1-7 and drug-related patents.2,4,8,9 Recent economic studies have recognized the government’s contributions to pharmaceutical innovation by contextualizing government as an “early-stage investor and government funding for research as an investment.”10-17 As such, these studies argued that there should be an equitable balance of investment risk and return between the public and private sectors15,16 and framed policy regarding the pharmaceutical industry’s drug pricing practices and profits as shaping this balance.17,18
The objective of this study was to compare NIH investment in the products approved by the US Food and Drug Administration (FDA) from 2010 to 2019 with reported levels of investment by the industry.19-21 This comparison required an accounting for NIH spending comparable with that used to estimate total industry investment. This typically includes not only costs associated with approved products, but also costs associated with products that fail in clinical development and the cost of capital, or opportunity cost, associated with these investments.19,20,22,23
Funding from the NIH for pharmaceutical innovation has been estimated from total NIH budget allocations24 or categorical funding from the Research, Condition, and Disease Categories or Research Portfolio Online Reporting Tools (REPORTER).25,26 These methods do not delineate spending associated with individual products. Case study methods have been used to identify NIH contributions associated with specific patents2,8 or products.5 These methods may not capture funding for basic research, which represents half of NIH funding and is classically undertaken “without specific applications towards processes or products in mind.”27
Other studies have focused on NIH funding for published research associated with approved drugs or their targets.6,28,29 In these studies, the costs of NIH-funded projects (grants) supporting research on a drug or its target were used as a measure of the NIH contribution to that product. In this method, drug-related publications represent applied research, and those associated with the drug’s target, but not the drug, represent basic research. Initial studies using this method identified NIH funding for research underlying each of the 210 drugs approved from 2010 to 2016, with total NIH costs of more than $100 billion and funding for each first-in-class drug of more than $800 million.28 These studies also demonstrated spillover effects in which NIH spending for basic research in immunology or endocrinology contributed to the development of products for treating cancer.29
This study extended these methods by developing an accounting for NIH spending that was comparable with reported investments by the industry. Using a data set of drugs approved from 2010 to 2019 (before the COVID-19 pandemic), this analysis estimated the NIH investment in these drugs, including the cost of published basic and applied research associated with these products, cost of phased clinical trials of failed product candidates, and opportunity cost, using discount rates recommended for government spending.30,31 These estimates were used to compare NIH and industry investments in new drug approvals, the cost savings to the industry provided by NIH spending, and the economic efficiencies created through spillovers of NIH-funded basic research on drug targets to multiple products. These results are discussed in the context of policy regarding drug pricing and corporate profit that affects the balance of investment risk and reward between the public and private sectors.
This cross-sectional study analyzed NIH-funding for published research related to drugs approved from 2010 to 2019 or their biological targets that was conducted from May 2020 to July 2022. This study did not involve human participants and was not subject to institutional review board review. The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
The core data collection of PubMed publications, NIH-funded projects and project costs associated with drugs approved from 2010 to 2019 has been previously described.6 Products approved by the FDA from 2010 to 2019 (new drug application or biologics license application [type 1]), excluding those derived from blood or tissue, diagnostic agents, vaccines, and antimicrobials, and dates of first approval were identified from annual FDA reports.32,33 Drug targets were identified from published literature34,35 or the Therapeutic Targets Database.36
Publications from 1960 to 2020 were identified in PubMed. Projects funded by the NIH from 2000 to 2020 were identified using the NIH REPORTER application programming interface. Projects were identified by NIH project number comprising the activity code, awarding institute, and number. Data on each project included the start year, end year, and costs for each fiscal year, subproject, or supplemental award. The analysis included phase-specific clinical success rates,19 average NIH costs for phased clinical trials,37 average industry investments,19,20 and drug-specific industry costs.20
Funding from the NIH for publications (PMIDs) associated with study drugs or their targets was identified in NIH REPORTER using methods described previously6,28 (eMethods in Supplement 1). Briefly, PMIDs were identified in PubMed using optimized search terms for drugs (eTable 1A in Supplement 1) or targets (eTable 1B in Supplement 1) as well as automatic term mapping protocols, including medical subject heading terms and Boolean modifiers. The PMIDs were indexed by PubMed Identifier, publication year, and search terms. Projects funded by the NIH that were associated with PMIDs were identified using the REPORTER publication link tables. The PMIDs were further associated with 1 fiscal year of project funding (project year) and total project costs for the year corresponding to the publication year. Project years and costs were not assigned to PMIDs published after the product’s first FDA approval, before the project start year, or more than 4 years after the project end year. Drug-specific costs were calculated from 2000 through the date of first FDA approval. To account for lags between funding and publication,38 PMIDs with publication dates 1 to 4 years after the project end year were associated with the project end year. The PMIDs identified by drug search were categorized as applied research, which included development. The PMIDs identified by target search, but not a drug search, were categorized as basic research. Project years and costs were categorized as applied research if 1 or more PMIDs associated with that project year were identified by drug search and categorized as basic research if none of the associated PMIDs were identified in drug searches. Duplicate PMIDs, project years, and costs were eliminated independently for each calculation.
The first drug associated with a novel biological target was classified as first to target.34,35,39 Applied research costs were estimated from costs identified through the drug search. Basic research costs were estimated from costs identified in searches for targets of first-to-target drugs. Averages were calculated after 95th percentile outlier elimination to account for searches with poor specificity. The average number of drugs per target was recalculated from Santos et al35 after excluding products derived from blood or tissue, diagnostic agents, vaccines, and antimicrobials. Spending from the NIH on failed clinical trials was estimated from phase transition rates19 and phase-specific NIH costs.37 Compounded 3% or 7% discount rates30,31 or a 10.5% cost of capital19 were calculated from the project year to first FDA approval.
Product-specific costs were compared for 81 first-in-class drugs with NIH costs estimated in this analysis and 63 drugs with industry costs described by Wouters20 using univariate regression in which Costi = β0 + β1 Sourcei in which Costi is the estimated NIH cost for research associated with the product or reported industry costs; Sourcei is an indicator variable with a value of 0 for NIH costs and 1 for industry costs; β0 estimates the median and 95% CI for NIH spending; and β1 estimates the median and 95% CI for the difference between NIH and industry spending. Costs were inflation-adjusted to 2018. Analyses were performed in Excel (Microsoft), PostgreSQL (PostgreSQL Global Development Group), or Python. All tests were 2 tailed. A 2-sided P < .05 was considered statistically significant.
The FDA approved 356 drugs from 2010 to 2019, including 336 associated with 217 known targets. PubMed searches for drug names identified 229 000 PMIDs, while searches for known drug targets identified 1.9 million publications, of which 21.4% had NIH funding (Table 1). Funding from the NIH funding was identified in 310 of 356 drug searches (87%) and in all 217 target searches (Table 1). Overall, this analysis identified NIH-funded research associated with 354 of 356 products (99.4%) approved from 2010 to 2019. The products without NIH funding were a chelating agent and osmotic laxative.
Funding from the NIH totaled $187 billion; $31 billion (17%) represented applied research on approved drugs, and $156 billion (83%) represented basic research on drug targets (Table 1). Figure 1 shows annual publications, NIH project (funding) years, and costs leading to first FDA approval.
Research projects and research-related programs, which typically support investigator-initiated research, provided 40.2% of NIH funding, including 43.2% of basic research costs and 24.8% of applied research costs. However, research program projects and centers as well as cooperative agreements (including clinical translational science awards), which typically contribute infrastructure or shared research capabilities, comprised 46.2% of total costs, 42.4% of basic research costs, and 65.6% of applied research costs (eFigure 1 in Supplement 1).
NIH Investment in Basic Research on Novel Targets
Of the 356 approvals, 86 (24.2%) were first-to-target products. Figure 1A shows NIH-funded publications, project years, and NIH costs associated with these targets leading to first-to-target product launch. Funding from the NIH was identified for all 86 targets (eTable 2 in Supplement 1).
Figure 2A shows NIH costs per novel drug target with no discount rate or 3% and 7% discount rates. After 95th percentile outlier elimination, the mean (SD) NIH cost for research on a novel drug target before a first-to-target product approval was $1.34 ($1.43) billion (3% discount, $1.63 [$1.74] billion; 7% discount, $2.15 [$1.66] billion; 10.5% cost of capital, $2.85 [$3.15] billion) (Table 2). Outliers included searches for CD-4, B-cell lymphoma 2, and epidermal growth factor receptor, which returned publications not explicitly associated with the drug target. Calculations without outlier elimination are shown in eTable 3 in Supplement 1.
NIH Investment in Applied Research on Drug Products
Figure 1B shows NIH-funded publications, project years, and NIH costs associated with applied research on 356 drugs through the year of approval. Before first approval, 301 of 356 products (84.5%) had NIH research funding (eTable 4 in Supplement 1). Figure 2B shows cumulative NIH costs for applied research with no discount rate or 3% and 7% discount rates. After 95th percentile outlier elimination, the mean (SD) NIH cost for applied research before approval was $51.8 ($96.8) million (3% discount, $58.5 [$111.9] million; 7% discount, $69.4 [$137.8] million; 10.5% cost of capital, $81.4 [$168.3] million) (Table 2). Outliers included searches failing to distinguish applied research on the approved product from basic research on the corresponding natural compound (ie, clotting factors, hormones, α-1 antitrypsin). Results without outlier elimination are shown in eTable 3 in Supplement 1.
Accounting for NIH Funding for Failed Product Candidates
The NIH costs calculation associated with failed clinical trials is shown in eTable 5 in Supplement 1. Based on reported phase transition rates,19 8.53 phase 1 trials, 5.08 phase 2 trials, and 1.79 phase 3 trials were conducted for each product approved. With NIH costs of $5.7 million for phase 1, $7.2 million for phase 2, and $3.9 million for phase 3,37 estimated NIH costs for clinical trials of failed candidates were $75.4 million for each product approval (3% discount, $80.6 million; 7% discount, $88.6 million; 10.5% cost of capital, $96.8 million) (Table 2).
Total NIH Investment to Launch First Drug Product Associated With Novel Targets
Total NIH costs were calculated for 86 first-to-target products as the sum of NIH costs for basic research on the target, applied research on the drug, and phased clinical trials of failed compounds. The distribution of costs is shown in Figure 2C. After 95th percentile outlier elimination, mean (SD) NIH costs before a first-to-target product launch was $1.44 ($1.37) billion (3% discount, $1.73 [$1.66] billion; 7% discount, $2.24 [$2.18] billion; 10.5% cost of capital, $2.96 [$3.11] billion) (Table 2). Data without outlier elimination are shown in eTables 2 and 4 in Supplement 1.
Comparing NIH and Industry Investments
DiMasi et al19 estimated average industry spending on 106 drugs approved from 1990 to 2010 at $1.5 billion or $2.8 billion with a 10.5% cost of capital (inflation-adjusted to 2018). Using different methods, Wouters et al20 reported an average industry spending on 63 drugs approved from 2009 to 2018 of $374.1 million, (95% CI, $301.9 million to $464.2 million) or $1.6 billion (95% CI, $1.27 billion to $1.89 billion) with a 10.5% cost of capital.
Spending from the NIH per approval for 81 first-to-target products was significantly greater than reported industry spending on 63 drugs20 before accounting for clinical failures, cost of capital, or discount rates (difference, −$1998.4 million; 95% CI, −$3302.1 million to −$694.6 million; P = .003) or with accounting for clinical failures (difference, −$1415.8 million; 95% CI, −$2731.4 million to $100.2 million; P = .04) (Table 3). Spending from the NIH was not less than industry spending when industry costs were estimated with clinical failures and a 10.5% cost of capital, and NIH spending was estimated with clinical failures and either a 3% discount rate (difference, −$1435.3 million; 95% CI, −$3114.6 million to $244.0 million; P = .09) or a 7% discount rate (difference, −$2436.3 million; 95% CI, −$4782.1 million to −$90.5 million; P = .04) (Table 3). Investment from the NIH and the industry was not significantly different when industry spending was estimated with clinical failures, prehuman costs19 (30.8% real costs), and a 10.5% cost of capital, and when NIH costs were estimated with clinical failures and either a 3% discount rate (difference, −$393.8 million; 95% CI, −$2120.5 million to $1332.9 million; P = .65) or 7% discount rate (difference, −$1394.8 million; 95% CI, −$3774.8 million to $985.2 million; P = .25) (Table 3).
Spillover Effects From Basic Research on Drug Targets
Santos et al40 cataloged 893 biological targets for FDA-approved products (1578) through June 2015, of which 1467 (93.0%) met inclusion criteria for this study. These products were associated with 515 biological targets, an average of 2.85 products per target (eFigure 2 in Supplement 1).
Accounting for spillovers of basic research on novel drug targets to 2.85 product approvals, the NIH cost for basic research per approval was $471.8 million (3% discount, $572.2 million; 7% discount, $753.4 million; 10.5% cost of capital, $1.0 billion) (Table 2). Accounting for spillover effects from basic research on drug targets, costs of applied research, product failures, and discount rates or cost of capital, the estimated NIH investment per approval was $599.0 million (3% discount, $711.3 million; 7% discount, $911.4 million; 10.5% cost of capital, $1179 million). Estimated NIH spending was lower than the reported average industry spending19 but within the 95% CI of per drug spending.20
In this cross-sectional study, evidence suggests the public sector makes substantial contributions to the foundational knowledge on which drug approvals are based,1,2,4,6-8,41,42 but less to patents6,9 or development.2,3,37,43 Conversely, the industry is primarily responsible for product development and sponsored more than 99% of the product launches in this data set.6
The objective of this work was to compare NIH investments in recent drug approvals with reported investment by the industry. This required an accounting for NIH spending with costs for basic research on the targets for these drugs, applied research on the approved products, phased clinical trials of failed products, and the recommended discount rates for government spending.30,31 This accounting adheres closely to methods used to estimate industry investment,19,20 while also recognizing fundamental differences in the nature of public and private sector investment in prevailing economic theories.10
These analyses suggest that NIH project costs for basic or applied research associated with the products approved from 2010 to 2019 were significantly greater than reported industry spending. Costs for the NIH were also higher than industry costs when both included spending on failed clinical trials of candidate products. Including clinical failures, NIH investment (calculated with either a 3% or 7% discount rate) was not less than industry investment calculated with a 10.5% cost of capital. Investment from the NIH calculated with clinical failures and a 3% or 7% discount rate was also not less than industry investment calculated with clinical failures, additional costs of prehuman research, and 10.5% cost of capital. These results suggest that NIH investments in pharmaceutical innovation are comparable with those made by industry.
While including the cost of capital in estimates of the industry’s investment in pharmaceutical innovation is controversial44 and estimates of this rate vary,45,46 consideration of the cost of capital is normative in finance theory and practice. These calculations are also consistent with prevailing economic theories that view private sector investment as inherently productive in that it typically generates a return on investment. In this context, the cost of capital represents the opportunity cost or financial risk that long-term capital investments in drug discovery and development may not achieve normal returns on investment.
There is no theoretical basis for applying an equivalent cost of capital to government spending. Prevailing economic theories treat government funding as nonproductive in that it is not expected to provide a return on investment. The 3% and 7% discount rates recommended by the US Office of Management and Budget for government spending30,31 have distinct theoretical foundations. The 3% discount rate on federal spending approximates the historical cost of government borrowing and, consequently, the full cost of government spending.31,47 The 7% discount rate represents the average productivity of private sector investments and is interpreted as a measure of the opportunity cost to the economy if public sector spending crowds out and reduces private sector investment.30,48 Given evidence that NIH funding for biomedical research stimulates, rather than reduces, private sector investment,49 estimating NIH investment with the 3% discount rate may be most consistent with prevailing economic principles.
This analysis did include an NIH spending calculation with the 10.5% cost of capital. This value provides an estimate of the additional costs that the industry would incur in the absence of NIH spending. Comparing these estimated cost savings with those of DiMasi et al19 or Wouters et al20 of industry investment suggests that industry costs would be more than double in the absence of the NIH contributions.
This work also recognizes that economic efficiencies may arise through spillover of knowledge or capabilities gained from NIH-funded basic research to applications by multiple firms or multiple products.8,29,35,43,50 Such spillovers would reduce the estimated NIH cost per approval. Considering only potential spillovers from NIH-funded basic research on drug targets to multiple products directed at the same targets, NIH spending per drug was within the range of actual industry spending.20 Spending from the NIH was estimated with either a 3% or 7% discount rate was lower than industry spending calculated with the 10.5% cost of capital.
Science and innovation policy remains grounded in a model in which government investments in basic research generate scientific capital that can be commercialized by industry for social and economic benefit. This model is exemplified by NIH spending for basic biomedical science, which plays an enabling role in pharmaceutical innovation.1,3,6,7,28,49 Emerging economic theory formalizes this model by contextualizing government funding for research as an early-stage investment in innovation.10-13,15-17 These theories further posit that, as early-stage investors, government or the public sector it represents could expect social or economic returns commensurate with those of comparable investments by the private sector.10,15,16
The present study was predicated on this concept that NIH spending represents an investment that can be meaningfully compared with investment by the industry. In this context, the finding that the magnitude of NIH investment in new drugs is comparable with that of the industry suggests that returns to the public and private sector should also be comparable. To achieve this, public policy associated with drug pricing,51 corporate profit,52 and commercial applications of government-funded invention53 should be calibrated to provide an equitable distribution of returns between the public and private sectors.10,15,16 The present results may provide a cost basis for considering not only the private rate of return to industry or the economy, but also the social return on investments,40 including the multiplex elements associated with health.54,55
First, this analysis is limited by the sensitivity and specificity of PubMed searches, right censoring of the data collection, and reported false-positive and false-negative associations between PMIDs and NIH projects in REPORTER.38 Search terms may not identify NIH funding for research tools, pharmaceutical modalities, or process development, which may underestimate total NIH costs.
Second, NIH costs for each publication were estimated as 1 fiscal year of project funding. This is consistent with evidence that 5-year NIH grants produce a median of 5 publications56 but may underestimate NIH costs for studies spanning multiple years.
Third, NIH funding in REPORTER represents a fraction of public sector funding for biomedical research and does not include funding from other agencies or governments, nongovernment organizations, academic institutions, or nonprofit organizations. This analysis also did not include contract funding, research and development tax credits, or vouchers. This would underestimate the public sector contribution to pharmaceutical innovation.
Fourth, this study considered only spillovers from basic research on drug targets. Spillovers may also emerge from NIH funding for research training, infrastructure, or capabilities. This would not affect the total NIH costs but would underestimate the gain from economic efficiencies.
This cross-sectional study found that NIH investment in drugs approved from 2010 to 2019 was not less than investment by industry, with comparable accounting for basic and applied research, failed clinical trials, and cost of capital or discount rates. The relative scale of NIH and industry investment may provide a cost basis for calibrating the balance of social and private returns from investments in pharmaceutical innovation.
Accepted for Publication: February 18, 2023.
Published: April 28, 2023. doi:10.1001/jamahealthforum.2023.0511
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2023 Galkina Cleary E et al. JAMA Health Forum.
Corresponding Author: Fred D. Ledley, MD, Center for Integration of Science and Industry, Bentley University, Jennison Hall 144, 175 Forest St, Waltham, MA 02452 (fledley@bentley.edu).
Author Contributions: Dr Ledley had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Ledley.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Ledley.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: All authors.
Obtained funding: Ledley.
Administrative, technical, or material support: Jackson.
Supervision: Ledley.
Conflict of Interest Disclosures: Drs Jackson, Galkina Cleary, and Ledley reported grants from the National Biomedical Research Foundation and Institute for New Economic Thinking during the conduct of the study as well as grants from the West Health Policy Center and National Pharmaceutical Council outside the submitted work. Dr Zhou reported grants from the National Biomedical Research Foundation during the conduct of the study as well as a grant from the National Pharmaceutical Council outside the submitted work. No other disclosures were reported.
Funding/Support: This work was supported by grants from the Institute for New Economic Thinking and the National Biomedical Research Foundation.
Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Data Sharing Statement: See Supplement 2.
Additional Contributions: We thank Juliana Harrison, MBA, Bentley University for assistance preparing the manuscript; Zoë Folchman-Wagner, PhD, and Vineeta Tanwar, PhD, Bentley University, for contributions to data collection; John Overington, PhD, VP Discovery Informatics, Exscientia and Olivier Wouters, PhD, London School of Economics, for consultation concerning their published works; and Michael Boss, PhD, Nancy Hsiung, PhD, and Bruce Leicher, JD, Bentley University, for their critical analysis. Drs Boss and Hsiung and Mr Leicher received no compensation for work at Bentley University or their role in this study.
3.Chakravarthy
R, Cotter
K, DiMasi
J, Milne
C-P, Wendel
N. Public- and private-sector contributions to the research and development of the most transformational drugs in the past 25 years: from theory to therapy.
Ther Innov Regul Sci. 2016;50(6):759-768. doi:
10.1177/2168479016648730
PubMedGoogle ScholarCrossref 6.Cleary
E, Jackson
MJ, Ledley
F. Government as the first investor in biopharmaceutical innovation: evidence from new drug approvals, 2010-2019. Accessed March 30, 2023. doi:
10.36687/inetwp133 10.Mazzucato
M, Li
H. The Entrepreneurial State: Socializing Both Risks and Rewards. Real-World Economics Review; 2018:84.
13.Block
FL, Keller
MR.
State of Innovation: the US Government’s Role in Technology Development. Routledge; 2015. doi:
10.4324/9781315631905 14.Hopkins
M, Lazonick
W. Who invests in the high-tech knowledge base? Accessed March 30, 2023. doi:
10.2139/ssrn.2638091 15.Lazonick
W, Mazzucato
M. The risk-reward nexus in the innovation-inequality relationship: who takes the risks? who gets the rewards?
Ind Corp Change. 2013;22(4):1093-1128. doi:
10.1093/icc/dtt019
Google ScholarCrossref 17.Lazonick
W, Hopkins
M, Jacobson
K, Sakinç
M, Tulum
ÖUS. Pharma's financialized business model. Accessed March 30, 2023. doi:
10.2139/ssrn.3035529 21.Rennane
S, Baker
L, Mulcahy
A. Estimating the cost of industry investment in drug research and development: a review of methods and results.
Inquiry. 2021;58:469580211059731.
PubMedGoogle Scholar 22.DiMasi
J. Causes of clinical failures vary widely by therapeutic class, phase of study.
Tufts CSDD Impact Report. 2013;15(5):1-4.
Google Scholar 37.Zhou
E, Jackson
MJ, Ledley
FD. NIH contribution to phased clinical development of drugs approved from 2010-2019: tech note. Accessed February 17, 2023.
42.Beierlein
JM, McNamee
LM, Walsh
MJ, Kaitin
KI, DiMasi
JA, Ledley
FD. Landscape of innovation for cardiovascular pharmaceuticals: from basic science to new molecular entities.
Clin Ther. 2017;39(7):1409-1425.e20. doi:
10.1016/j.clinthera.2017.06.001Google ScholarCrossref 44.Chit
A, Chit
A, Papadimitropoulos
M, Krahn
M, Parker
J, Grootendorst
P. The opportunity cost of capital: development of new pharmaceuticals.
Inquiry. 2015;52:0046958015584641.
PubMedGoogle Scholar 46.Harrington
SE. Cost of capital for pharmaceutical, biotechnology, and medical device firms.
The Oxford Handbook of the Economics of the Biopharmaceutical Industry. eds, Danzon
PM, Nicholson
S. 2012:75-99. doi:
10.1093/oxfordhb/9780199742998.013.0004 49.Toole
AA. Does public scientific research complement private investment in research and development in the pharmaceutical industry?
J Law Econ. 2007;50(1):81-104. doi:
10.1086/508314
Google ScholarCrossref 54.Lakdawalla
DN, Doshi
JA, Garrison
LP
Jr, Phelps
CE, Basu
A, Danzon
PM. Defining elements of value in health care-a health economics approach: an ISPOR Special Task Force report [3].
Value Health. 2018;21(2):131-139. doi:
10.1016/j.jval.2017.12.007
PubMedGoogle ScholarCrossref 55.Stiglitz
JE. Measuring What Counts: The Global Movement for Well-Being. The New Press; 2019.