Clinical Research Has System Sprawl Problem

By John Oncea, Chief Editor, Clinical Tech Leader

When it’s working as designed, technology can be phenomenal. Systems flow, teams flow, and, ultimately, trials flow. But when there’s a glitch? Systems slow to a crawl, teams get frustrated, and trials can fail.

Some technological failures are more annoying than anything: a printer jamming, a system update at an inconvenient time, or a wearable failing to sync. Then there are more serious technological failures – failed EDC systems, inadequate audit trails and data review tools, and automation errors – that ramp up frustration levels to dangerous levels.

Then there are the types of problems that are maddening, and there may be no better example of this than system sprawl, the uncontrolled proliferation of IT infrastructure, software, and applications, leading to complex, fragmented, and underutilized environments.

System sprawl is, unfortunately, highly prevalent in clinical research, as evidenced by this National Center for Biotechnology Information (NCBI) report finding that more than half of research sites are struggling to manage a multitude of disconnected, sponsor-provided technologies.

The NCBI study further found that 59% of trials are hindered by the use of multiple disconnected systems, causing increased manual data re-entry (68%) and inefficient workflows (58%).

What about high site burden? Sites often manage six or more different logins and systems per study, a number that increases with complex oncology trials that sometimes require more than 15 systems.

The prevalence of silos? Ninety-seven percent of organizations report at least one major challenge with their clinical applications, with 68% citing the difficulty of integrating multiple systems.

This system sprawl has operational consequences, from fragmented collaboration to duplicated work to elevated data quality risks, and is getting worse, with 67% of surveyed stakeholders reporting set up and training on sponsor technology as more burdensome today than five years ago.

That’s simply unsustainable, and left unaddressed, system sprawl will only get worse.

System Sprawl Is Creeping Normality

To understand why the industry keeps tolerating it, you have to understand how it starts.

Pulitzer Prize-winning author Jared Diamond coined the term “creeping normality” in his 2005 book, Collapse: How Societies Choose to Fail or Succeed. As illustrated by the boiling frog fable, creeping normality is a process by which a major, often catastrophic, change is accepted as normal because it happens slowly, in small, unnoticeable increments.

This is how system sprawl in clinical research almost always happens: slowly and incrementally. It is a progressive, cumulative shift triggered by a rise in decentralized, hybrid, and digital trials that accelerated the adoption of specialized tools, resulting in a complex landscape of systems, including EDC, IRT, ePRO/eCOA, and various portals. Add in fragmented purchasing, a lack of standardization, and increased study complexity (more data, more endpoints, and decentralized components increase the sheer volume of data-generating tools), and BANG! You’re a boiled frog.

Florence’s 2026 State of Clinical Trial Technology Report, based on the results of a survey of more than 400 clinical trial professionals – including 326 site leaders, 46 sponsor executives, and 44 CRO decision makers – further illustrates the problems system sprawl is causing:

• 35% of sites use 3–4 internal systems daily, while 37% use 3–4 study-specific systems per active trial
• Some sites report using 9 or more internal systems and 9 or more study-specific systems for a single study, resulting in up to 18 systems per study
• 64% of sites report that sponsor-provided technology duplicates their existing workflows, leading to parallel documentation and manual reconciliation
• Over half of the sites report a neutral to negative effect on startup time and staff efficiency due to system complexity

System sprawl is a significant challenge in clinical trial technology, leading to inefficiencies, duplication of work, and operational friction. ​Addressing this issue requires improved system integration, workflow connectivity, and a shift toward a unified digital infrastructure.

Fighting System Sprawl With The Right Technologies

The clinical research industry has not been short on diagnoses when it comes to system sprawl. The harder question always has been treatment. Fortunately, a convergence of technologies, standards, and regulatory pressure is beginning to offer something the field has long lacked: a credible path from fragmentation to coherence.

Building A Common Language

The most immediate lever is interoperability, and the backbone of that effort is the HL7 Fast Healthcare Interoperability Resources (FHIR) standard, which enables disparate electronic health record systems and clinical trial platforms to exchange data using a common language. FHIR is not new, but its clinical research applications are maturing rapidly. Researchers have demonstrated that EHR data can be extracted via FHIR and used to pre-populate electronic case report forms directly, eliminating the kind of redundant manual entry that currently consumes site staff time across studies.

The CDISC and HL7 joint FHIR-to-CDISC Mapping Implementation Guide, published through the Journal of the Society for Clinical Data Management, formalized the bridge between healthcare data standards and research data standards, making it technically feasible to move patient data from an EHR into a trial workflow without transcription errors or format conflicts.

That bridge matters because the sprawl problem is, at its root, a data problem. When every sponsor-provided system speaks a different language, sites are forced to become translators, keying the same information into six, ten, or fifteen different platforms per study. Shared standards remove the translation burden. CDISC’s Digital Data Flow initiative, currently in its fourth phase of public review, is building exactly that kind of shared foundation: a standardized, machine-readable study definition that can flow downstream into EDC systems, IRT platforms, and regulatory submissions without being re-entered at each handoff. The governing principle, as framed by researchers writing in Applied Clinical Trials, is straightforward: create data once, use it everywhere.

Regulations Catching Up To Reality

The regulatory environment is increasingly aligned with this vision. The finalization of ICH GCP E6(R3) in January 2025, with its explicit support for digital health technologies and eSource data, represents the most significant modernization of Good Clinical Practice guidelines in decades. The guideline added a data governance section covering the full data life cycle, signaling that regulators now expect sponsors and sites to think systemically about data flow rather than platform by platform.

Technology vendors have taken notice. The industry is moving, with varying degrees of speed and conviction, away from point solutions toward what researchers in the Journal of Medical Internet Research describe as unified, integrated, and trial-specific digital platforms. These environments consolidate EDC, eCOA, eConsent, patient engagement, and regulatory document management into a single governed architecture, replacing the patchwork of logins and manual reconciliation with a single source of truth. The efficiency argument is not abstract: the same JMIR research notes that few commercial platforms are currently open to integrating third-party components, and that the absence of universal data standardization remains a limiting factor. The technology exists; adoption is the variable.

AI As Integration Engine

Artificial intelligence is emerging as a powerful accelerator of integration, particularly in the area of data quality and workflow automation. A 2025 study published in PubMed by researchers at the Welingkar Institute of Management and Amity University found that AI-driven automated validation, real-time monitoring, and anomaly detection significantly reduce errors across the trial process while improving regulatory compliance and transparency. Many organizations are also adopting AI for data standardization, automating the normalization of terminology across therapeutic areas so that data collected in one system can be meaningfully compared with data collected in another, reducing one of the core frictions that makes sprawl so operationally costly.

Bring Your Own Technology

Looking immediately ahead, the concept gaining traction most rapidly is what the industry is calling Bring Your Own Technology. Developed through a collaboration between the Decentralized Trials and Research Alliance and site leaders from organizations including GSK and Mayo Clinic, BYOT inverts the traditional sponsor-imposed tools model.

According to Clinical Leader, instead of every sponsor layering a new system on top of whatever a site is already running, BYOT allows sites to use their own validated eClinical systems, with interoperability standards like CDISC ODM and FHIR maintaining the data visibility sponsors require. The practical effect is that a well-resourced academic medical center, for instance, no longer needs to train staff on a different technology stack for every incoming study. The number of logins drops; the quality of data entry improves; the burden on-site personnel decreases.

Applied Clinical Trials describes the broader shift as “platformization,” a transition in which traditional EDC and protocol management tools give way to dynamic, machine-readable living protocols and automated data capture. The vision is a trial environment that is not just digitized but genuinely self-integrating: one in which a protocol amendment propagates automatically through connected systems rather than requiring a cascade of manual updates across a dozen vendor portals.

None of this is without friction. Legacy systems resist replacement, smaller sites lack the resources to invest in new infrastructure, and cultural resistance to change runs deep in an industry accustomed to document-centric operations. But the direction of travel is clear. The sprawl did not accumulate overnight, and it will not be resolved overnight either. What is different now is that the technical vocabulary for integration, the regulatory framework for digital data, and the platform architecture for consolidation are all converging at the same moment. The boiled frog has finally noticed the temperature.