By Deborah Borfitz 

November 11, 2025 | It is no secret that clinical trial protocols have grown increasingly complex, resulting in longer study durations, higher costs, and greater burden on patients and investigative sites. It has been a consistent, well-documented trend for more than two decades now, according to Kenneth Getz, executive director of the Tufts Center for the Study of Drug Development (Tufts CSDD), who has made it part of his life’s work to quantify the problem. 

Most recently, Tufts CSDD teamed up with TransCelerate BioPharma in a common quest to deliver the evidence needed to bring about long-awaited and lasting remedies. Their collaboration reveals that not all the data sponsors are collecting in their clinical trials supports each study’s primary and key secondary endpoints, says Getz.  

“With all the data we’re collecting, we are also making it harder and harder for sites and patients to participate in our studies,” he adds. “We’re really trying to nail down this idea that protocol design has to be more intentional and ... meaningful.” 

Optimizing data collection practices isn’t just about creating efficiencies but also reducing the burden on sites and study participants to enable needed medicines to get to market as quickly as possible, stresses Jeneen Donadeo, executive director of portfolio management for TransCelerate. Surprisingly, the latest study done with Tufts CSDD suggests that obstacles to achieving this goal may be common across companies and therapeutic areas, indicating industry-wide opportunities to optimize data collection and improve inefficiencies. 

Phase 3 protocols, remarkably, have hit the 6 million data point mark, Getz says, even though the proportion of data being collected by non-core procedures has declined over time. The good news is that clinical teams and protocol authors were able to identify core, standard, and required procedures that were not essential for assessing the key objectives and endpoints crucial for a clinical trial’s success. 

The non-essential procedures are study-specific, he continues. Their commonality is related to collecting data on patients more often than necessary; for example, on visits 6 and 11 in addition to visits 1, 2, 4, and 5, which would have been sufficient to demonstrate the primary or key secondary endpoints.  

Quantifying the magnitude of non-core and non-essential procedures in protocols has revealed opportunities to take some of the onus off study sites and participants through smarter, more streamlined data collection, says Getz. Despite mounting pressures on sponsors to collect more data, ICH E6 (R3)—the new “roadmap” for Good Clinical Practice—Donadeoexplicitly supports fit-for-purpose data collection even as it recognizes that protocols are becoming more customized in both a scientific and executional sense. 

Room for Cutbacks 

Companies have many strategic reasons for collecting data in study protocols that have nothing to do with a drug’s primary and key secondary endpoints, says Getz. They might want to gather more data that could potentially be part of a product label, to address the interests of payers and providers once drugs are introduced into the marketplace or to gain insights into future use of the medicines or into biomarkers that can shed light on the disease and its progression. 

This is on top of global regulatory requirements—some country-specific—demanding more data collection in clinical trials to ensure compliance, research integrity, and patient safety, he adds. The complexity of diseases being studied as well as new methodologies like virtual and remote approaches being used to support protocol execution are contributing factors. 

Key findings of the latest collaborative study suggest there is ample room for cutbacks. Nearly one-third of all procedures and associated data collected in clinical trials (40% for non-oncology studies) were found to be scientifically unnecessary, representing at least a quarter of the total burden on patients and sites. Findings were drawn from 105 phase 2 and 3 protocols across 14 biopharmaceutical companies, all of which are TransCelerate members. 

Moreover, over 70% of the non-core procedure data ultimately appeared in the clinical study report. “The primary purpose for collecting that data is for future use and exploratory purposes and to support insights into the safety of the molecule or of the therapy,” says Getz. “This shows that there is perceived strategic value in the data collected and a desire to include some of this tertiary, exploratory, and miscellaneous data in the CSR” even if it is not relevant to demonstrating the key objectives and endpoints of the study.  

His focus is on the newfound connection between all the data being gathered, and the burden it represents to sites that must administer the procedures and participants who must adhere to protocol requirements. This is where there are opportunities to simplify protocol study design and reduce the unnecessary, if unintended, impacts, Getz says. 

Working Smarter 

The study was a monumental undertaking and would have been impossible without the “commitment and perseverance” of TransCelerate member companies, Getz points out. The data collection process alone took about nine months because everyone had to be trained on how to consistently code the data, and companies were given a “warmup period” in which to practice and get feedback. 

Donadeo saw it as a hopeful sign that so many companies opted to participate in the study. Sponsors recognize that change is needed but are driven by hard evidence to make it happen, she says. 

“As someone who has been directing teams to collect this type of data and do these types of studies for more than two decades, I can definitely recognize and sense how hard it is to apply our findings and insights to meaningfully change protocol design behavior,” says Getz. External pressures coupled with the difficulty of managing the interests of multiple functions within companies “almost offsets the gains that are made from some of the earlier studies that we’ve conducted.” 

But regulatory tailwinds support a closer examination of trial-related activities and the associated endpoints that are being supported as studies have become more customized for narrowly defined patient populations often dealing with difficult diagnoses including cancer and rare diseases, he continues. This means looking at the burden that is being created for sites and participants as trials are designed to ensure they include the most relevant activities, in sufficient amounts, to support the primary and key secondary endpoints. 

The required behavior change represents a “mindset shift” away from just-in-case data collection practices and begins with “thinking about the purpose of every data point ... early in the protocol design process,” says Donadeo. Sometimes, that might mean getting regulatory input on what is truly needed to answer a research question. 

But critical decision-making is also needed at the upfront design stage, which is why TransCelerate will be putting together “considerations frameworks” as well as helping its members share current best practices that have a track record with regulators. The end game is “to make smarter data collection the industry norm and not the exception,” she says.  

‘Journey of a Data Point’ 

To raise awareness of how data overload occurs and identify opportunities for improvement, one initial initiative of TransCelerate is to simply follow the “journey of a data point,” Donadeo shares. “I don’t think all the stakeholders always understand where it goes and what the implications are of making certain decisions along the way.”   

“In our discussions with the participating companies, fit-for-purpose [data collection] also implies tradeoffs that have to be made,” adds Getz. It’s an acknowledgment that certain types of procedures are “nice to have but they also have a downstream impact on site and participant burden.” 

The latest study with TransCelerate revealed a couple of protocol design areas, where a run-up in non-essential or non-core procedures is typically seen, which are ripe for targeting, Getz says. Questionnaires and patient diaries often see a compounding number of questions, for example, creating a “demanding experience” for study volunteers. 

Since protocols are so highly customized, it could be tricky to create best practices for questionnaires or any other feature of study design. So, TransCelerate has instead landed on the idea of composing a high-level checklist of potential areas to explore early in the design process, says Donadeo. The work will fall to the Optimizing Data Collection team, which will have subteams devoted to different deliverables that could include regulatory engagement.  

Casting the Vision 

Socialization and communication of potential strategies for optimizing the collection of protocol data is already well underway, Getz says. In September, findings of the collaborative study were shared in a keynote presentation at the DPHARM: Disruptive Innovations to Advance Clinical Trials Conference in Boston. And earlier this month, he moderated a keynote panel at SCOPE Europe with four big-pharma executives on ways to close the gap between protocol development and feasibility to ensure downstream success with sites and patients. 

The study manuscript is publicly accessible as a pre-print, and has been submitted for publication to Therapeutic Innovation & Regulatory Science. “We’re hopeful that the paper, at least the online peer-reviewed version, will be available before the end of the year,” he says. 

Industry sponsors, including many smaller organizations, have already expressed a lot of interest in the findings, says Getz. Companies are particularly appreciative of the non-essential procedures variable—defined as “procedures determined by the clinical team or protocol authors as being performed in excess of the number of times required to demonstrate a clinical outcome”—which is entirely new. 

This is not the first time Tufts CSDD and TransCelerate have teamed up, but the latest study was the largest and longest running collaborative project to date. The credibility and know-how of Getz and his team to derive real-world insights from member companies made them logical partners in the shared mission to drive “meaningful, scalable change,” says Donadeo.  

Only six TransCelerate members were unable to participate, she adds, and not for a lack of interest. For some, the project was simply too resource-intensive to prioritize, in part because such a large amount of data was being requested within a time threshold. 

A “more mindful data collection mindset” could reap significant benefits, even with a tiny reduction in the overall volume of information being amassed in a study, says Donadeo. “More data isn’t always better. Every procedure, every data point really should be intentional and aligned with the study’s goals if we really want to improve trial efficiency ... [and] reduce the burden for patients and sites.”