Avoiding Redundancy in Clinical Trials With Proactive Artificial Intelligence

Avoiding Redundancy in Clinical Trials With Proactive Artificial Intelligence

As of November 2022, 434 thousand clinical trials had been registered, approximately 35,000 more than 2021. This growth shows no signs of slowing down, and the proliferation of clinical data that these trials produce is driving innovation and drug discovery. That’s all to the good, but it also creates an urgent problem for the pharmaceutical industry at large: with so many trials ongoing, how do companies avoid duplicating research that has already been done?

In this article, we will discuss the costs and risks of redundant or duplicate clinical trials, and propose a much-needed solution for the problem of research waste. 

What exactly is a redundant clinical trial?

As the costs of clinical trials continue to balloon, cutting down on research waste should be an urgent priority for the entire industry. Regulators in many countries agree: China’s top regulator recently warned that it would be taking a harder line on redundant trials that waste resources. 

Broadly speaking, a clinical trial is considered redundant if the question it poses can or has been resolved on existing evidence. There is an important distinction between this kind of trial, phase IV trials that investigate side effects of treatments that have already entered the market. Phase IV trials are essential to assessing drug safety, and are, of necessity, justified by a review of existing data. Redundant trials, on the other hand, are not.

A significant 2020 study found, alarmingly, that half of Randomized Clinical Trials (RCTs) did not cite a Systematic Review (SR) before commencing the trial. This oversight radically increases the risk of research waste, with dire financial and ethical consequences.

The costs of redundant trials: the obvious and the hidden

The first and most obvious cost of wasted research is that of conducting the clinical trial itself. This includes all activities from planning to execution:

  • pre-study preparation and writing protocols
  • preparing for enrollment and recruiting participants
  • conducting the study, administering interventions and collecting data
  • analyzing results and writing reports
  • obtaining regulatory approval for any products used during the clinical trial

In addition, further costs emerge after the trial: auditing by an independent third party, and the costs of marketing new products when they’re ready. The critical point here is that, for a truly redundant trial, all of this can and should have been avoided. The biggest cost of all, lurking behind all of the ones we have listed, is the opportunity cost incurred by pouring resources into dead-end research. That may be difficult to quantify, but it can be mission-ending.

Costly and harmful: a lose-lose for the pharmaceutical industry and patients alike

Perhaps more importantly, redundant trials expose patients themselves to unnecessary and unjustifiable risk. Risk is inherent to all clinical trials, and justifying that risk is an integral part of the systematic review process. Where that process is inadequate, or omitted altogether, human life and well-being are jeopardized. Recent research from China demonstrated the human cost of wasteful research: in one case, thousands of adverse events and hundreds of deaths. As regulators around the world move to close in on redundancy, companies can expect to face more scrutiny – and more penalties – for wasteful research in future.

Leveraging AI to safely navigate oceans of data

The way out of this crisis is obviously not to halt or even slow down innovation. What’s needed is more intelligent innovation that can avoid duplicating research and wasting resources. To achieve that, R&D professionals need a way to accurately survey existing data and identify genuine white spaces. But the rapid proliferation of data points has already made traditional methods of monitoring technical data obsolete. 

Artificial Intelligence is the solution that the industry desperately needs, and it holds the promise of revolutionizing the way research is conducted in the life sciences. Traditionally, pharmaceutical R&D teams are forced to rely on labor-intensive research practices that they simply cannot scale without drastically increasing personnel numbers. Artificial Intelligence is now able to replace much of that workflow, proactively scanning millions of data points and presenting them to the humans who need to make critical business decisions. 

Know everything you need to know, moment to moment, with Similari

At Similari, we’ve brought together industry leading expertise from AI, machine learning and NLP to create exactly the kind of innovation intelligence that enables pharmaceutical researchers to reliably identify what’s been done before, what’s currently being done, and where they can move in the future. 

To learn more about how to future-proof your innovation projects with cutting-edge data surveillance and monitoring tools, schedule a demo with the Similari team. We’re always ready – just like our technology – to show how much humans can achieve when they have all the information they need.

Related Articles


Expensive Failures: 3 Drug Development Missteps, and...

After years of development and billions in R&D spend, the vast majority of drugs never make it to market. Failure is a...

March 29, 2023 | 5 min read

Read More


The Anatomy of a Search Query: How...

The word “Google” became a verb in 2006, when it found its way into the Merriam-Webster dictionary. But like all shifts in...

March 29, 2023 | 5 min read

Read More


Planning Your 2023 Conference Schedule: So Many...

In-person conferences are back in full swing, and the event schedule for pharma and related industries is already jam-packed. But with so...

March 18, 2023 | 5 min read

Read More