Clinical trial costs go under the microscope

The FDA wants to rein in the expenses of these studies, but that could prove challenging.

Search for this author in:

Glow Images/ Getty

The safety of the drugs we use hinges on the rigorous testing of these compounds in a clinical setting. But now more than ever, testing a new medical therapy requires deep pockets. Global spending on clinical trials is estimated to reach $68.9 billion a year by 2025, according to Grand View Research, a market-research company. There are efforts emerging to rein in growing costs. In September, the US Food and Drug Administration Commissioner Scott Gottlieb said he wanted to reduce the cost of clinical trials through various methods, including the use of seamless trials, which save money by running continuously instead of in three phases. Still, some say that the current system is too dysfunctional to fix and that only a complete overhaul will do.

How much does a clinical trial cost? The answer to that question is tricky. The cost depends on the methods used and the disease, among other things. For example, studies that use surrogate markers, such as cholesterol in cardiovascular trials, can cost less than ones that rely on clinical outcomes, such as survival in cancer trials. Thomas Moore, a senior scientist for drug safety and policy at the Institute for Safe Medication Practices (an educational healthcare nonprofit), and his colleagues reported in November in JAMA Internal Medicine1 that the median cost of a clinical trial between 2015 and 2016 was $19 million. They added, “At the extremes of the distribution were 100-fold cost differences.” So the range is wide, to say the least.

Costs are particularly high for late-stage clinical trials, which often involve large cohorts of patients. According to Gen Li, president of Phesi—a company that claims to have the “industry’s most comprehensive and dynamic clinical trials database”—analysis of a sample of 4,363 phase 3 trials revealed an average cost of “around $30 million to implement” each.

The location of a trial also impacts its cost2. For example, in China or India, a clinical trial can be conducted for a cost that is 30–40% lower than it would be in the United States—for various reasons, including lower prices of testing procedures, such as an electrocardiogram. But the outsourcing of clinical trials to other countries is not without controversy.

One thing that most experts will agree on is that the cost of clinical trials is increasing. Solid numbers support that claim. For example, The US Bureau of Labor Statistics stated that the mean cost per patient in a clinical trial increased more than fourfold—rising from $3,773 in 1989 to $16,567 by 2011. This report noted an average annual growth rate of 7.5% in this metric.

Someone might suspect that higher costs create better results—perhaps because more advanced technology that produces more accurate data costs more—but that’s not necessarily true. “A clinical trial in 2019 will cost double what it cost in 2008, and with no increase in the quality of data or the speed of acquiring it,” says Tomasz Sablinski, CEO at Transparency Life Sciences.

Complex causes

When asked about the top reasons behind rising clinical trial expenses, Sablinski says, “The one-word answer is complexity.” And, seemingly everything—patient enrollment, trial designs and data analysis—is complex in most of today’s trials. During his 25 years in the pharmaceutical industry, including more than a decade at Novartis, he’s watched that complexity increase. “Most of the complexity,” he says, “is absolutely unnecessary.”

Beyond cost increases from complexity, Sablinski says, “In the early ‘90s, with the exponential growth of biotech, it became a very price-insensitive industry.” With companies hoping to market a multi-billion-dollar drug, he explains, it doesn’t matter if a trial costs $10 million or $50 million. That triggers a chain reaction of cost increases.

“People who make money on clinical trials are incentivized to charge a lot,” Sablinski says. As an example, he points out contract research organizations (CROs)—of which his company is one—and says that they are “very economically driven to make clinical trials more expensive.” Of course, Sablinski is trying to do just the opposite. Plus, it’s not just the CROs. Sablinski adds that the price of services in a clinical trial, say a magnetic resonance imaging (MRI) scan, is often 2–4 times the typical market cost. So, a $2,500 procedure for an ordinary patient might cost $10,000 for a clinical-trial patient.

Even the compound being tested can have a big impact on price. “You can test things like aspirin and buy it economically in kilos,” Li says, “but using patient-specific antibodies involves processes that make the drug extraordinarily expensive.”

Bringing costs down

With such trial-dependent costs, it’s more difficult to pinpoint the best way to reduce overall expenses. “It won’t happen with any one single thing, and it won’t happen in the near term,” says Lisa Henderson, group editorial director at Applied Clinical Trials & Pharmaceutical Executive. Nonetheless, she sees possibilities for economizing, such as identification of potential patients through genetic tests and other molecular markers. With that, plus some telemedicine and other web-based methods of assistance, says Henderson, “eventually, the costs of a supported infrastructure would likely go down.”

Sablinski agrees on the value of adding methods like web-based processes, but that’s precisely what his company does. Still, he says that running an operation that’s as digital as possible can “reduce complexity by creating a situation where you reduce the amount of transactions needed to get the data from a patient into a database, digitizing it through telemedicine and wearable devices.” By comparison, he points out that banking 40 years ago involved going to a branch and talking to a teller, and now it’s mostly digital. Current clinical trials are more like banking 40 years back, but Sablinski says that making these more digital could reduce the cost of most trials by 20–30%.

In addition, Li believes that picking the right investigators to participate makes trials more efficient. “The idea of getting a good investigator is extraordinarily complicated,” he says. Using statistical models and dynamically tracking the performance of investigators in clinical trials, Li and his colleagues maintain a database of top choices. Li adds that the “design of a clinical trial and the business process to deploy the investigators are critically important."

The reality

Coming up with ideas for reducing the cost of clinical trials will be easier than instituting them. “The current system is entrenched,” Sablinski says. “It will take a tremendous battle to change it, because it involves so many groups and so many people interested in making money.”

Making incremental improvements over time is surely simpler to do, but it’s not guaranteed to create an efficient and economical system. Sablinski says more drastic measures are needed, from the use of crowdsourcing in designing trials to digital tools to stratify patient populations. “The system has to be completely rethought, and that’s what we are doing,” he concludes.

Anyone hoping to tear down and rebuild a system that drives a market heading quickly toward $65 billion a year faces a gargantuan task. On the other hand, with overall US health care eating up close to 20% of the GDP, and drugmakers citing clinical trial costs as one driver behind pharmaceutical prices, this is one battle that must be fought.

doi: 10.1038/d41591-019-00008-7


  1. 1.

    Moore, T. J. et al. JAMA Intern. Med. 178, 1451–1457 (2018).

  2. 2.

    Ali, S. et al. Int. J. Clin. Trials 5, 194–199 (2018).

Download references

Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up