Olivier J. Wouters · JAMA 2020 · retrospective economic cost analysis · n=63

Estimated Research and Development Investment Needed to Bring a New Medicine to Market, 2009-2018

Cited 1539 times in the scientific literature.

Level 4 - case-series / case-control

Level 4 by non-clinical design analogy (retrospective economic cost analysis of public financial records and regulatory data).

OpenAlex W3009999522 · doi:10.1001/jama.2020.1166 · record verified 2026-08-29

What was done

Researchers analyzed publicly available financial and clinical development data for new therapeutic agents approved by the US Food and Drug Administration (FDA) between 2009 and 2018. Data were gathered from US Securities and Exchange Commission filings, Drugs@FDA, ClinicalTrials.gov, and published trial success rates. Total capitalized R&D investment was estimated in 2018 US dollars incorporating costs of failed trials and an annual 10.5% cost of capital.

What was found

Of 355 FDA-approved drugs and biologics during the study period, R&D expenditure data were available for 63 products (18%) developed by 47 companies. In the base-case analysis accounting for trial failures, the median capitalized R&D cost was $985.3 million (95% CI, $683.6 million–$1228.9 million), and the mean was $1335.9 million (95% CI, $1042.5 million–$1637.5 million). By therapeutic category (with ≥5 drugs), median costs ranged from $765.9 million (95% CI, $323.0 million–$1473.5 million) for nervous system agents to $2771.6 million (95% CI, $2051.8 million–$5366.2 million) for antineoplastic and immunomodulating agents.

Why it matters

This study provides an empirical estimate of drug development costs based entirely on public regulatory and financial filings, offering a transparent benchmark that is lower than several prominent industry estimates but shows substantial variation across drug classes.

Limits

R&D expenditure data were accessible for only 18% (63 of 355) of approved drugs, creating potential selection bias toward smaller companies, orphan drugs, first-in-class agents, and accelerated approvals. Estimates also relied on modeling assumptions regarding preclinical spending, cost of capital, and overall clinical trial success rates.

Cited by