By Deborah Borfitz

February 18, 2025 | In a debate at the recent Summit for Clinical Ops Executives (SCOPE) over whether fraud and misconduct in clinical trials represent a marginal nuisance or a major problem, the argument for it being a serious situation easily won the day based on a show of hands in the room. Making the case for an embarrassment of breaches was Jonathan Rowe, head of R&D quality, operations and risk management at management consulting and technology firm ZS. 

Rowe was engaged in a friendly clash with his colleague Marcin Makowski, head of centralized monitoring and data analytics at GSK, who playfully held strong to his stance that “the 98% is more important than the 2%.” Makowski borrowed the comment reportedly made by a principal investigator downplaying the relative impact of fudging a few numbers.  

The first to speak was Rowe, reporting that 80% of submissions to the drug authority in China were thrown out after the launch of a self-assessment program a decade ago. The suggestion is that fraud is rampant in some places and “perhaps it’s a good idea to oversee it,” he says. 

Makowski counters that “drawing conclusions” requires reliable and representative data, which in this case is hard to ascertain. More importantly, he says, “to say something is happening often and to say it has a lot of meaningful consequences” are not the same.  

A physician by training, Makowski says it is nonsensical to measure the respiratory rate in someone who is not in respiratory distress, although it is common to take such measurements on everyone participating in a clinical trial. When asked to do so as a medical student, he shares, he would fabricate a figure, such as 22 (a bit above the normal respiratory rate), since it had no apparent meaning. 

In the world of clinical research, real problems are being overshadowed by concerns about fraud and misconduct, says Markowski. Consider the oncology product that failed to translate from accelerated to full approval for a company in late 2023 despite technically positive results of an open-label study.  

The basis of the drug’s removal from the market by the advisory committee was that around 12% of patients who were to get chemotherapy withdrew their consent right after learning of their assigned treatment arm. There was no reported fraud and no protocol deviations, since “patients have the right to withdraw consent at any moment,” Makowski notes. 

So, the trial failed for reasons other than fraud, which happens, Rowe retorts. A quick search on the topic of “clinical trials and fraud” on Google Scholar turns up 99,000 hits, which suggests a problem even if only 40,000 of them are on point. Many of the papers are published by biostatisticians in the industry, based on algorithms unique to different companies. Fraud is “leading to a waste in resources and rework and expenses minimally,” says Rowe. 

Makowski readily agrees that there is intentional manipulation or fabrication of data collected during clinical trials, but that control methods exist to discover and take care of it. “It’s not a problem because it’s easily policed,” he says. 

“At least put the police car on the highway... so people know they are being watched,” counters Rowe, returning to his China example where “people knew they were not being watched. What would happen if our investigators and our sites and the city of Miami knew that nobody was paying attention? Would fraud become rampant?” 

Question of Intentionality 

Fraud comes with “reputational risk,” Rowe says. If a key opinion leader is the offender and works with a company’s medical affairs group and commercial team as well as on clinical trials that breach can spread “up the ladder into other aspects of how people behave.” 

In the ICH E6(R3) Guideline on Good Clinical Practice, the terminology referencing fraud is “potential data manipulation,” he says. The document writers “want you as a monitor and a sponsor to identify significant errors in data collection” but don’t take a stand on whether it is or is not a rampant practice.  

There isn’t necessarily a “causal relationship” between what’s happening and the public’s perception of the situation, which is subject to influence by one’s favored news channel, says Makowski. All in all, the attention paid to potential misconduct is disproportionate to the impact on the results of a study. “Sometimes, the impact on the study comes from the things that were done wrong because they cannot be done right and not because somebody is making things wrong intentionally.” 

Many companies have experienced fraud-related quality events, says Rowe, and the consequences for the trial may not be readily apparent. “There are a lot of things we don’t know we don’t know,” including—as he has seen—investigator training being handed off to a clinical research associate. 

The literature on fraud is skewed heavily toward studies done within academia rather than by pharma companies, Rowe says. “We as an industry are also giving medical grants for research and, sometimes, we’re making decisions off of what comes out of these investigator-run grants.” Those funded trials can’t always be replicated in-house. 

Most articles about misconduct are cases happening in academic preclinical basic research, says Makowski, because the researchers rely on grants and their livelihood is dependent on getting published. Those events cannot be extrapolated to clinical research, especially late-phase trials.  

The question is whether these fraudulent academic researchers are the same individuals companies work with in their clinical trials, Rowe says, suggesting many of them indeed are. Fraud-related waste and rework are in any case a costly problem for industry. He cites the case of the vaccine trial where several thousand patients had to be withdrawn due to violations of Good Clinical Practice at certain clinical trial sites.  

The Audience Speaks

Makowski found a fan with one debate watcher pointing to the “plane crash theory” whereby bad news like the vaccine study and fraudulent behaviors in China get more attention than the many untold trial success stories. Rowe says he lectures on the topic, which he refers to as the “shark attack theory,” pointing out that falling airplane parts are in fact more deadly than a news-making shark bite.  

Among those in Rowe’s corner was a nurse, saying fraud and misconduct are rampant because they’re not viewed as such by the perpetrators. She cites cancer studies where oncology centers tend to follow their own regimen for dosing calculations rather than what the protocol stipulates. “Those are all protocol deviations that at the end of the day can significantly impact the data that we’re making our decisions on,” she says. 

A former quality lead at a sponsor organization also reports that she has come up against a few cases where, after a site had already been paid, a death certificate was found with a date prior to when a study visit was reported to have been completed on the electronic case report form. One academic investigator voiced his indifference to a mere 2% of a trial using this flavor of fraud.  

For her, it was a loss of trust. Even if the trial data for that one study were unaffected, misbehavior can get normalized in multiple other ways in the future.  

“Fraud has ramifications,” echoes another Rowe supporter, and jeopardizes multiple companies who might end up using the same investigator. “For us, data is data integrity and if you don’t have that, I’m sorry, we won’t use you again.” 

“I left big pharma five years ago, and at the time we were hesitant to share information about sites across the industry,” says Rowe. But when fraud is detected, it might make sense for sponsors to share the news with one another so offending sites can start facing the consequences.