The Integration Dilemma: Who Decides Which Tech Stack Wins?
By John Oncea, Chief Editor, Clinical Tech Leader
The first article in this series established that sites are not the obstacle to technology adoption in clinical trials; they’re often ahead of where sponsors assume. The deeper problem begins when both sides have systems in place, and neither is willing to simply abandon them. That collision – whose platform takes precedence, whose workflow gets reshaped, and who bears the cost – is the integration dilemma.
It doesn’t have a tidy answer. But understanding why it persists, and where the realistic path forward lies, requires looking honestly at the incentives driving each side of the conversation.
This article is the second of a three-part series based on the Clinical Tech Leader Live event, Managing The Tech Stack In Today's Trials. Click here for part one, and click here for part three.
Why Sponsors Say No, And What It Takes To Change That
Joe Dustin of Dauntless eClinical Strategies was direct about the sponsor-side calculus: getting a yes requires more than a good idea. A site that wants to plug its own technology into a sponsor’s infrastructure needs to bring validation documentation up to sponsor-level standards. Without it, the answer is almost always no, not out of indifference to site burden, but because sponsors see risk without commensurate return.
“If the benefit of the site’s workload is the only return, sponsors aren’t going to do it,” Dustin said. “There has to be a benefit for them as well: data quality, speed to data entry, reduction in query and deviation cycle times, less monitoring time.”
That framing matters. Integration cannot be positioned as charity extended to sites. It has to function as a business case: faster, cleaner data that reduces sponsor overhead. For academic sites running older custom systems, the lift required to meet validation standards can be substantial, and sponsors may not see enough upside to justify the investment.
The implication is that the conversation needs to change before the technology can. Sites that want sponsor buy-in have to be able to articulate their value proposition in sponsor terms, not just site terms.
Economics Are More Flexible Than They Appear
That said, Rosie Filling of Keenova Therapeutics pushed back on the assumption that the economics are fixed. Sponsors – at least mid-size ones – are often more open to cost-sharing than sites realize. If a site’s technology reduces burden, improves recruitment, and supports cleaner protocol execution, there’s a business case on both sides. And if sites don’t ask, they’ll never find out what’s negotiable.
“It doesn’t hurt to ask,” Filling said, specifically in response to a participant question about whether sponsors might help fund site-facing technology. In her view, that question should be routine. If a platform is being used across multiple studies and generating measurable value, a sponsor may be willing to split the cost, including startup costs for sites that are earlier in their technology journey.
That reframes the economics from a barrier into a negotiation. Not every integration has to be a zero-sum exercise. But that conversation has to happen early – before the site has already absorbed the cost – and it requires sponsors to accept that they don’t always know what sites already have or what they need.
The Communication Gap Is Upstream
Beth Harper of the Tufts Center for the Study of Drug Development identified the root of many integration failures: sponsors often assume they already understand site technology, so they don’t ask. Sites interpret that silence as disinterest. Neither side gets the information it needs.
Harper’s prescription is specificity. “Tech” means different things to different people. A site might have an EMR that it doesn’t use for research. Asking whether a site is “digitally enabled” produces noise, not signal. The useful questions are concrete: Does the site have eConsent? ePayment? ECOA? eLab portals? An eSource platform? Those answers tell you what actually exists and what can realistically be connected.
Filling extended the point to the site side: even when sponsors aren’t asking the right questions, sites shouldn’t wait. Sites should be proactively communicating their capabilities, in feasibility surveys, during pre-study visits, wherever the channel exists. Sponsors can’t work with what they don’t know about.
The Standards Problem, And Why Custom Work Doesn’t Scale
Dustin identified a structural problem beneath the integration dilemma: the industry still behaves as though every trial is a bespoke project. That mindset makes sense at the individual study level, but it destroys scalability. A sponsor running a Phase III trial across 200 sites cannot build 200 separate integrations. The only path to scale runs through standards, and the industry has a complicated history with those.
FHIR was designed as a transport standard; it’s been pushed into use cases it wasn’t built for. CDASH was meant to standardize EDC inputs; it’s routinely customized in ways that undercut its purpose. Every time the industry bends a standard to fit a specific trial, it makes the next integration harder. Regulatory mandates – like the requirement for SDTM datasets – show what’s possible when the industry is forced to align. The question is whether that kind of alignment can happen proactively rather than reactively.
Dustin also noted a shift already underway at the vendor level. Newer companies entering the clinical technology market are building API-first, interoperability-first, standards-first architectures from the ground up. Legacy systems, built primarily to satisfy regulatory requirements rather than to communicate across platforms, are increasingly hard to integrate with. The backend shift happening among newer vendors matters more to long-term scalability than any front-end feature.
Whose Stack Actually Wins?
The honest answer is that no single stack wins in a landscape this diverse. Filling returned to a core principle: sites are not interchangeable, and sponsors are not in a position to mandate a universal workflow. Sites working with dozens of sponsors simultaneously can’t rebuild their operations for each one. And as she pointed out bluntly, sponsors need sites and participants more than sites need any individual sponsor. Sites with resource constraints increasingly choose which studies to take on. Sponsors who ignore that dynamic risk making their own trials harder to staff.
The practical path forward isn’t about picking a winner. It’s about building enough common infrastructure – validated, interoperable, economically viable for both sides – that integration becomes the default rather than the exception. That means fewer custom workarounds, more proactive communication, and a shared willingness to ask better questions before the technology gets deployed.
Once the industry accepts that integration will never be one-size-fits-all, the next challenge becomes building a practical tech stack that actually reduces site burden without adding new layers of complexity. That’s where the conversation heads in the third article of this series, Building a Clinical Trial Tech Stack That Actually Works.