From 73% Noncompliant To 2.4%: Building A Compliance Office
By John Oncea, Chief Editor, Clinical Tech Leader
When Yolanda Davis joined the University of Miami in 2013, the institution’s clinical trial disclosure noncompliance rate stood at 73 percent. Two years later, it was 2.4 percent. The turnaround happened under the university’s Research Compliance and Quality Assurance (RCQA) Office, where Davis built the Clinical Trial Disclosure (CTD) unit from the ground up, and it’s the kind of specific, dated, numbers-driven result that’s become rare to find in clinical research operations coverage.
A 73 percent noncompliance rate on clinical trial disclosure isn’t a minor administrative gap; it’s a regulatory exposure problem touching the majority of a university’s active trials. It’s also tempting to read a number that high as a training problem or an accountability problem. Davis saw something different: institutions rarely arrive at 73 percent noncompliance because people are ignoring the rules. More often, they lack standardized workflows, defined ownership, and reliable tracking, which makes the failure an infrastructure problem wearing a compliance label.
Davis built CTD within RCQA specifically to address it, standing up the processes, tracking, and accountability structures that hadn’t existed before.
The Two-Year Turnaround, Year By Year
The improvement wasn’t a single event; it happened in measured stages. In the first year, noncompliance dropped from 73 percent to 13 percent. In the second year, it dropped again, from 13 percent to 2.4 percent. That year-over-year detail matters more than the headline number: it shows a deliberate, sustained process rather than a one-time cleanup, and it's the kind of granular, first-hand data point that only exists because Davis lived through building it.
The results were significant enough that Davis went on to speak at FDA conferences about the work, translating what she’d built at one institution into guidance other institutions could use.
Building A Tool Other Institutions Now Use
Within the CTD build, Davis created a tool that the University of Miami has since posted publicly. Other institutions have reached out asking to license it or learn how it works, an unusual outcome for something originally built to solve one university’s internal compliance problem.
It’s a useful counterpoint to a later admission Davis makes: getting institutional buy-in for a purchased, vendor-built system was far harder than getting people to adopt something built in-house to solve a problem they already recognized. It raises a question worth sitting with for any technology leader: is resistance aimed at the tool itself, or at a solution to a problem people don’t yet believe they have?
From Compliance Office To Investigator-Initiated Trials
Davis ran CTD from 2013 to 2020. In 2021, she moved on to build a second office: the Clinical Research Management and Support Office, since renamed Investigator-Initiated Trial Services at the U (IITS-U). That unit supports multicenter, investigator-initiated trials, and its portfolio has tripled since 2021.
The two builds required different approaches, by Davis’s own account. Building CTD inside an already-established institution meant leading with documented process, then bringing people into that structure that already existed. Standing up IITS-U as a newer unit meant reversing that order entirely: getting the right people in place first, then building formal process around them as the unit grew, rather than asking a process to hold together before there was a team to run it.
There was no single playbook that carried over between the two builds, just a consistent instinct for which sequence a given environment actually needed. That distinction tracks with a broader point Davis makes about organizational maturity generally: the right starting point depends on whether you’re fixing something established or standing up something new, and treating both the same way is one of the more common mistakes she sees institutions make.
What She’d Do Differently
Asked what she’d change, Davis points to timeline planning: she didn’t originally build in enough time for change adoption, which created friction that slowed early progress on both offices. The technical work, in other words, moved faster than the human side of it.
Resistance from leadership, she says, was best handled by reframing proposed changes in terms of financial exposure and institutional reputation rather than process for its own sake, shifting the conversation from “why do we need this” to “what happens if we don’t.” And she treated pushback as “not right now” rather than “never,” revisiting ideas later rather than abandoning them outright when the timing wasn’t there.
That patience is easy to overlook in a case study built around a clean before-and-after number, but it’s arguably as much a part of the 73-to-2.4 story as the process work itself: sustained institutional change, in Davis’s experience, moves at the speed leadership is willing to be persuaded, not the speed a project plan says it should.
Thirteen years, two offices built from scratch, and a noncompliance rate turned from a serious liability into a model other institutions now study. For a Florida State graduate who has spent over a decade building infrastructure at Miami, one of FSU’s longtime rivals, it’s a career that’s outlasted the joke.